/* ---- Google Analytics Code Below */

Thursday, December 31, 2020

Robots Dancing, Today

 From Boston Dynamics, now part of Hyundai.  At the link they outline each of the dancers backgrounds. Less scary than some views.  Or is it? 

Whats Next in Tech

An outline from HPE

Exploring What’s next in tech – Insights, information, and ideas for today’s IT and business leaders

Accelerating the future.  The race to stay ahead in an increasingly digital world

Organizations are digitally transforming at an unprecedented pace. Here's what you need to know.

Report - 4.40 MB - 40 Pages

We explore all that's needed to make your digital transformation successful—from strategies for your workforce to how to ensure your tech is secure.

One undeniable effect of the pandemic is the sudden embrace of digital transformation. Enterprises of all types and sizes have sped up their digital plans to meet with the new reality: remote workforces, remote customers and partners, and just about everything needing to run digitally. Leadership teams everywhere have pulled projects off the planning schedule and put them into high gear.

This issue of The Doppler is designed to help you navigate all that involves, from how to establish and achieve your main goals to ensuring that your systems are set up in the most secure way possible. We also reconsider the principles and guideposts of digital transformation itself, a worthwhile topic considering the various ways that people use the term digital transformation.

What you will find inside:

Is your company achieving business outcomes or following the herd?

Here's how to execute a successful digital transformation strategy with the digital journey map.

Five strategic patterns for successful digital transformation

A new study gives guidance businesses can reuse and adapt as they pursue their digital strategies.

Transform your customer experience to unleash your company's full potential

Practical insights for companies trying to disrupt their own business models before someone else does.

Don't go it alone. Three successful strategies for making partnerships work.

Successful innovation in the digital economy increasingly requires an ecosystem of partners.

The reimagined workplace

Predictions and recommendations for the future of work.

Zero trust makes business secure by default

As enterprises turn to hybrid IT, they find that conventional identity and access management doesn't keep up.  ... '

Wednesday, December 30, 2020

Amazon Buys Wondery for?

 Though most recently have been looking through podcast offerings, had not heard of Wondery.    Note too that Amazon now also has the talking book system Audible, which I have also used.   All these deal with the interaction of voice and text, which made me think.   In general podcasts are very useful in delivering (usually) narrow chunks of information.  Podcasts are interesting, but the way they are delivered can be difficult to use effectively. I also found Audible also hard to use in that way.   Is this an attempt to pull to pull the broader idea of textual knowledge together and more easy to use?   Or as the article suggests, a means of connecting content providers?  More like the Spotify direction also mentioned?    I like the idea of immersive knowledge delivery.  

Amazon buys Wondery, setting itself up to compete against Spotify for Podcast Domination

Wondery is now part of Amazon Music

By Ashley Carman@ashleyrcarman in theVerge

Wondery, one of the last major independent podcast networks, is now owned by Amazon. The companies announced the deal today, ending speculation about who might eventually buy the network, which is most well-known for true crime podcasts like Dirty John. (Apple and Sony were also reported to have at least discussed a possible purchase.)

Amazon didn’t disclose the acquisition price, although earlier reports from Bloomberg and The Wall Street Journal suggested Wondery’s value was at least $300 million. Wondery is technically now part of Amazon Music, which launched podcast support in September this year.

The acquisition is a notable one, if only because it sets Amazon up to better compete against Spotify, which has been acquiring networks and exclusive talent for over a year. Amazon says this Wondery deal doesn’t change existing shows — they’ll still be available through “a variety of providers.” But Amazon says it hopes buying Wondery will “accelerate the growth and evolution of podcasts by bringing creators, hosts, and immersive experiences to even more listeners across the globe.” ... ' 

Review Manipulation in Amazon

 Had only just recently directly noticed fake or manipulated reviews on Amazon.  Had thought that Amazon was controlling these, but then alterted to it by an ArsTechnica article, see below.  Remember being impressed by 'extensive detailed  reviews' on Amazon, especially for complex technical items.    When I buy something expensive or complex I am likely to be cautious in its selection.   But examples show in some cases clearly not even reviewing the product at hand.  Inserted to skew statistics in comparisons.    It cheapens the whole idea of reviews as a useful measure.   Certainly AI could detect these by patterns using AI.  Lets fix it. 

Amazon still hasn’t fixed its problem with bait-and-switch reviews

Amazon has done little to stop sellers from gaming its search algorithm.

By Timothy B Lee ... in ArsTechnica

City Researchers Reveal Link

Seems there would be a link,  but what kind of behavior is indicated?  Always looking for predictive behavior we can leverage.

City Researchers Reveal Link Between Coding of Cryptocurrencies and Their Market Behavior

City University of London (U.K.)   By John Stevenson

A connection between the coding of cryptocurrencies and their market behavior has been identified by researchers at the U.K.'s City University of London. The researchers studied 297 cryptocurrencies whose code is stored in GitHub, and whose average daily trading volume during their lifetime was higher than $100,000. They found that 4% of developers contribute to the code of two or more cryptocurrencies, raising questions about the transparency of the coding process. City University's Andrea Baronchelli said, "Cryptocurrencies are open source digital objects traded as financial assets that allow, at least theoretically, everyone to directly shape both an asset structure and its market behavior. Our study, identifying a simple event in the development space that anticipates a corresponding behavior in the market, establishes a first direct link between the realms of coding and trading."

Tuesday, December 29, 2020

Schneier on SolarWinds

A considerable piece, with much supporting general and technical information and links, on the SolarWinds attack.  Also includes many comments I am working my way through. 

Russia’s SolarWinds Attack  by Bruce Schneier

Recent news articles have all been talking about the massive Russian cyberattack against the United States, but that’s wrong on two accounts. It wasn’t a cyberattack in international relations terms, it was espionage. And the victim wasn’t just the US, it was the entire world. But it was massive, and it is dangerous.

Espionage is internationally allowed in peacetime. The problem is that both espionage and cyberattacks require the same computer and network intrusions, and the difference is only a few keystrokes. And since this Russian operation isn’t at all targeted, the entire world is at risk — and not just from Russia. Many countries carry out these sorts of operations, none more extensively than the US. The solution is to prioritize security and defense over espionage and attack. ...... " 

Formerly in the Guardian. 

737: Management and Technology

My daughter in law flies for American Airlines, has been on furlough for awhile now.   The most recent issue of the ACM outlines elements of how 737 technology has been managed.   All technology is by definition managed, and this aeronautical-technical piece is an excellent example. 

Boeing's 737 MAX: A Failure of Management, Not Just Technology  By Michael A. Cusumano   Communications of the ACM, January 2021, Vol. 64 No. 1, Pages 22-25  10.1145/3436231

Neural Networks are Bayesian

 Just reading this, worth thinking about. Technical.

Neural networks are fundamentally Bayesian  in TowardsDataScience By Chris Mingard

Stochastic Gradient Descent approximates Bayesian sampling

Deep neural networks (DNNs) have been extraordinarily successful in many different situations — from image recognition and playing chess to driving cars and making medical diagnoses. However, in spite of this success, a good theoretical understanding of why they generalise (learn) so well is still lacking.

In this post, we summarise results from three papers, which provide a candidate for a theory of generalisation in DNNs [1,2,3].     .... "

Voice AI Trends in Europe

Podcast and more of work being done in and for Europe.  From Voicebot

Voice AI in Europe Year-in-Review 2020 with 169 Labs, Soapbox Labs, and Voxalyze – Voicebot 

Podcast Ep 185  Bret Kinsella in Voicebot

oday we talk about voice AI adoption and trends in Europe for 2020. My guests include Niamh Bushnell, chief communications officer of Soapbox Labs, Dominik Meissner, co-founder of 169 Labs, and Alexis Hue, founder and managing director of Voxalyze.

The guests provide a broad perspective given that Soapbox offers ASR tech for children in custom voice interactive experiences for education and entertainment, 169 Labs develops voice experiences for consumer brands and enterprises, and Voxalyze has a SaaS solution for voice app publishers looking to grow their audience.

Today we discuss privacy, whether there is a voice assistant duopoly, custom assistants, voice and media, personalization, voice app developer disappointment, the impact of the COVID-19 pandemic, and much more all from a European perspective.   ... '

Detecting Sarcasm

Even in the early days of reading consumer reactions to product communications  we knew it was useful to have an idea of how much sarcasm was being used.  To determine human reactions to statements or images.   Often we just scrubbed potential sarcasm. but we knew we were losing something by not analyzing it.

This work on Detecting Sarcasm  also Discussed in  VentureBeat  

Modeling Intra and Inter-modality Incongruity for Multi-Modal Sarcasm Detection

By Hongliang Pan, Zheng Lin, Peng Fu, Yatao Qi, Weiping Wang


Sarcasm is a pervasive phenomenon in today’s social media platforms such as Twitter and Reddit. These platforms allow users to create multi-modal messages, including texts, images, and videos. Existing multi-modal sarcasm detection methods either simply concatenate the features from multi modalities or fuse the multi modalities information in a designed manner. However, they ignore the incongruity character in sarcastic utterance, which is often manifested between modalities or within modalities. Inspired by this, we propose a BERT architecture-based model, which concentrates on both intra and inter-modality incongruity for multi-modal sarcasm detection. To be specific, we are inspired by the idea of self-attention mechanism and design inter-modality attention to capturing inter-modality incongruity. In addition, the co-attention mechanism is applied to model the contradiction within the text. The incongruity information is then used for prediction. The experimental results demonstrate that our model achieves state-of-the-art performance on a public multi-modal sarcasm detection dataset. ... 

Monday, December 28, 2020

Making Smart Thermostats More Efficient

 Beyond the simplest adaptable method.

Making Smart Thermostats More Efficient

MIT News

Researchers from the Massachusetts Institute of Technology (MIT) and Russia's Skolkovo Institute of Science and Technology (Skoltech) collaborated on the design of a smart thermostat that uses data-efficient algorithms to learn optimal temperature thresholds within seven days. The thermostat's algorithms use reinforcement learning and are event-triggered, making decisions only in response to events defined by specific conditions reaching a threshold. Skoltech's Henni Ouerdane said, "A learning thermostat can potentially learn how to adjust its set-point temperatures in coordination with other HVAC devices, or based on its prediction of electricity tariffs in order to save energy and cost."

On the Geopolitics of AI

Useful thoughts on power and balance, beyond AI, but also all kinds of analytical method.

The geopolitics of artificial intelligence

24 Dec 2020| By Anastasia Kapetas

As artificial intelligence technologies become more powerful and deeply integrated in human systems, countries around the world are struggling to understand the benefits and risks they might pose to national security, prosperity and political stability.

These efforts are still very much a work in progress. Australia is developing a whole-of-government AI action plan, led by the Department of Industry, Science, Energy and Resources. The department released a discussion paper  this year and finalised its call for submissions in November.

In line with the department’s brief, the paper concentrates on the economic potential of AI, while acknowledging the need for a human-centred, ‘responsible’ AI regime. That reflects a push internationally to conceptualise AI in terms of digital human security and human rights.

But AI technologies also have serious implications for national security and geopolitics, which need to be thoroughly explored in any discussion of what an AI framework for Australia might look like.

In any discussion of AI, it’s important to note that definitions of the technology are not settled and the applications are vast. But most definitions circle around the idea of machine learning, the ability of a digital technology not just to automate a function, but to learn from interactions with its environment and optimise a function accordingly.

The AI systems that we need to think about in national security terms include surveillance and profiling, the persuasive AI that pervades digital social networks, predictive algorithms and autonomous systems. It is also important to think about the control of the entire AI supply chain, from the human source of the datasets that AI technologies need to learn from, to research and development and technology transfers, and the effects of AI systems on societies.  ... " 

Doing Simple Chatbots

Close to the kind of thing we did simply in the 90s.  In the Alexa Blog:

How to Create a Chatbot With No Coding Experience in 4 Simple Steps  By Lydia Roth

If you really want to drive customer engagement, learning how to create a chatbot is a great place to start.

According to Dmitriy Kachin, COO of Chatfuel, strong chatbot experiences with higher audience engagement register 80–90% response rates. This result shows that users are willing to use chatbots, but the experience still needs to be properly executed to meet their needs.

Building a chatbot completely from scratch using custom code takes a lot of time and resources that many businesses simply don’t have. Lucky for you, we’ve broken down a step-by-step process for quickly and cost-effectively building a chatbot — no coding experience necessary.

Step 1: Define the Goal of Your Chatbot

Before you actually start building your chatbot, you need to set a clearly defined goal for it. The type of chatbot you build will depend on what you want it to accomplish.

Goal Example 1: Handle a Portion of Incoming Customer Support Questions

Want to make life a little easier for your hardworking support staff? Use a chatbot to resolve simple inquiries so your staff can focus on more complex issues that require extra attention and detail.

Use a chatbot to resolve simple inquiries so your staff can focus on more complex issues that require extra attention and detail.   ...  " 

Broader Strategies for Cyberattack

 Being more precise about detection data gathering and retaliation.   Wondering what the context of the existing modeling included. 

A Better Kind of Cybersecurity Strategy

MIT News   By Peter Dizikes

Researchers at the Massachusetts Institute of Technology (MIT), Northwestern University, and the University of Chicago contend Russia's use of North Korean IP addresses for a cyberattack during the opening ceremonies of the 2018 Winter Olympics underscored the need for a new cybersecurity strategy involving selective retaliation. Said MIT's Alexander Wolitzky, "If after every cyberattack my first instinct is to retaliate against Russia and China, this gives North Korea and Iran impunity to engage in cyberattacks." After extensive modeling of scenarios in which countries are aware of cyberattacks against them but have imperfect information about the attacks and attackers, the researchers found a successful strategy involves simultaneously improving attack detection and gathering more information about the attackers' identity before retaliating. Wolitsky added, "If you blindly commit yourself more to retaliate after every attack, you increase the risk you're going to be retaliating after false alarms."

DeepMind Masters Games without the Rules, Just Rewards

MuZero. An intriguing  example, with the right data.

 DeepMind's AI Masters Games Without Even Being Taught the Rules

IEEE Spectrum by   Philip E. Ross

The MuZero artificial intelligence (AI) developed by Alphabet subsidiary DeepMind can master games without learning their rules beforehand. The system attempts first one action, then another, learning what the game rules permit, and concurrently noting the rewards proffered. MuZero then modifies its strategy until it finds a way to collect rewards more readily; this observational learning is perfect for any AI confronting problems that cannot be specified easily. MuZero outperforms previous DeepMind AIs in its economical use of data, because it models only environmental parameters that matter in decision-making. Once trained, MuZero processes so little for decision-making that its entire function might be managed on a smartphone. ... "

Pandemic Holiday Retail Post Mortem

What Just Happened?  Good overview by a number of experts on retail process.  Join us in January for a recap of the craziest holiday season on record.  Wed., Jan. 27 – 12 noon ET/9 am PT

(Can't make it? Register and we'll alert you when the recording is available.)  Register Now:

All attendees will get a copy of the full research report. Join us live for the analysis, panel discussion and audience Q&A.

Amanda Nichols  Sr. Manager, Industry Marketing, UKG   Rob Snyder  COO & Co-founder, SYRG, Bob Phibbs  President/CEO,  The Retail Doctor, Moderating:  Al McClain  CEO, Co-founder, RetailWire

Grab a virtual seat for this holiday post-mortem of seasonal hiring, staffing and in-store safety, based on a just-completed pulse survey of retail managers conducted by UKG, with additional analysis from the workforce gurus at SYRG. You'll get reactions to the findings and advice as well in our panel discussion featuring Bob Phibbs, The Retail Doctor — insights that will help retailers streamline and optimize workforce operations in the year ahead. 

We’ll do a prior-year comparison to reveal:

How stores changed their approach to seasonal hiring and staff engagement;

How associate performance impacted store success;

To what extent COVID-19 had a direct impact on store operations (and staff morale), and;

Whether managers deemed their stores fully prepared for the holiday rush.   ... 

Sunday, December 27, 2020

Modifying the Way Companies Talk

Because machines are analyzing the patterns in what is said.   And of course, machines will soon also generate some of the conversation.

How machines are changing the way companies talk   By Khari Johnson @kharijohnson  in VentureBeat

Anyone who’s ever been on an earnings call knows company executives already tend to look at the world through rose-colored glasses, but a new study by economics and machine learning researchers says that’s getting worse, thanks to machine learning. The analysis found that companies are adapting their language in forecasts, SEC regulatory filings, and earnings calls due to the proliferation of AI used to analyze and derive signals from the words they use. In other words: Businesses are beginning to change the way they talk because they know machines are listening.

Forms of natural language processing are used to parse and process text in the financial documents companies are required to submit to the SEC. Machine learning tools are then able to do things like summarize text or determine whether language used is positive, neutral, or negative. Signals these tools provide are used to inform the decisions advisors, analysts, and investors make. Machine downloads are associated with faster trading after an SEC filing is posted.

This trend has implications for the financial industry and economy, as more companies shift their language in an attempt to influence machine learning reports. A paper detailing the analysis,    originally published in October by researchers from Columbia University and Georgia State University’s J. Mack Robinson College of Business, was highlighted in this month’s National Bureau of Economic Research (NBER) digest. Lead author Sean Cao studies how deep learning can be applied to corporate accounting and disclosure data.

“More and more companies realize that the target audience of their mandatory and voluntary disclosures no longer consists of just human analysts and investors. A substantial amount of buying and selling of shares [is] triggered by recommendations made by robots and algorithms which process information with machine learning tools and natural language processing kits,” the paper reads. “Anecdotal evidence suggests that executives have become aware that their speech patterns and emotions, evaluated by human or software, impact their assessment by investors and analysts.”

The researchers examined nearly 360,000 SEC filings between 2003 and 2016. Over that time period, regulatory filing downloads from the SEC’s Electronic Data Gathering, Analysis, and Retrieval (EDGAR) tool increased from roughly 360,000 filing downloads to 165 million, climbing from 39% of all downloads in 2003 to 78% in 2016.

A 2011 study concluded that the majority of words identified as negative by a Harvard dictionary aren’t actually considered negative in a financial context. That study also included lists of negative words used in 10-K filings. After the release of that list, researchers found high machine download companies began to change their behavior and use fewer negative words. ... ' 

ReThinking the Global Supply Chain

Yes, do a better job of risk analyses.

Is It Time to Rethink Globalized Supply Chains?

The COVID-19 pandemic should be a wake-up call for managers and prompt them to consider actions that will improve their resilience to future shocks.

By Willy Shih  in SloanReview

The COVID-19 contagion has had a major impact on Chinese manufacturers, and because of the central role many Chinese companies play in the supply chains of other companies, the impact is being felt around the world. The disruption is particularly acute in the electronics and auto industries, but it is also affecting pharmaceuticals, metals, and a wide range of consumer and industrial products, including surgical gowns and masks.

How did we end up with such complex interdependency in our supply chains — and what should managers be thinking about once we get through this?  ... 

Coders Flock Back to Crypto Projects

 Intriguing direction.  Of the belief that developers will become replaced by automated methods

Coders Flock Back to Crypto Projects with Prices Surging Again

In Bloomberg  By Olga Kharif

Software developers are migrating back to cryptocurrency projects at record levels, with early-stage venture firm Electric Capital reporting that the numbers of new coders rose this year for more than three consecutive months for the first time since 2017. Electric estimated the number of new crypto developers grew 15% per month during the first 10 months of 2020, while more than 80% of all active developers began their work in the last two years. Open source blockchain platform Ethereum and decentralized finance applications are particularly attractive to developers. Electric's Maria Shen said, "Many projects like smart contract platforms rely on developers joining their ecosystem to be successful. Ethereum clearly has more activity and this is why other platforms fight for developers.”

Saturday, December 26, 2020

Transmitting Holograms for Remote Experience

Do we need a hologram machine in every home?  For telemedicine and work?   Replacing our zoom stadia?   So we might have crowds of realistic holograms in situ?  Is it necessary, and what is really necessary at all anymore?  Now much improved,  since seeing them first at Disneyworld.  Would be impressive, am in for the idea.

PORTL Hologram raises $3M to put a hologram machine in every home  By Jonathan Shieber@jshieber in TechCruch

What does a hologram-obsessed entrepreneur do for a second act after setting up a virtual Ronald Reagan in the Reagan Memorial Library, or beaming Jimmy Kimmel all the way from Hollywood to the Country Music Awards in Nashville?

If that entrepreneur is David Nussbaum, the founder of PORTL Hologram, the next logical step is to build a machine that can bring the joy of hologram-based communication to the masses.

That’s the goal thanks to a new $3 million round that Nussbaum’s company raised from famed venture investor Tim Draper,  former Electronic Arts executive Doug Barry and longtime awards-show producer Joe Lewis.

Barry is not only backing the company, he’s also coming on board as its first chief operating officer.

Much of this interest can be traced back to the hologram performance given posthumously by Tupac Shakur back at Coachella about eight years ago.

Nussbaum turned the excitement generated by that event into a business. He bought the patents that powered Tupac’s beyond-the-grave performance, and used the technology to beam Julian Assange out of the Ecuadoran embassy he had been holed up in during his years in London and making dead stars live (and tour) again.

Those visual feats were basically just an updated version of the Pepper’s Ghost technique that stage illusionists and moviemakers have been using since it was invented by John Pepper in the 19th century. ... ' 

Secure Multi Party Computation

 A considerable discussion of the problem in the most recent issue of the ACM.  In general the notion of Multi Party Computation (MPC) was something I had not thought about directly,  but since my background is about computation, this is worth thinking through.

Secure Multiparty Computation   By Yehuda Lindell

This article also contains a video at :  bit.ly/3nD1DDm

Communications of the ACM, January 2021, Vol. 64 No. 1, Pages 86-96   10.1145/3387108

Distributed computing considers the scenario where a number of distinct, yet connected, computing devices (or parties) wish to carry out a joint computation of some function. For example, these devices may be servers that hold a distributed database system, and the function to be computed may be a database update of some kind. The aim of secure multiparty computation is to enable parties to carry out such distributed computing tasks in a secure manner. 

Whereas distributed computing often deals with questions of computing under the threat of machine crashes and other inadvertent faults, secure multiparty computation is concerned with the possibility of deliberately malicious behavior by some adversarial entity (these have also been considered in the distributed literature where they are called Byzantine faults). That is, it is assumed that a protocol execution may come under "attack" by an external entity, or even by a subset of the participating parties. 

The aim of this attack may be to learn private information or cause the result of the computation to be incorrect. Thus, two important requirements on any secure computation protocols are privacy and correctness. The privacy requirement states that nothing should be learned beyond what is absolutely necessary; more exactly, parties should learn their output and nothing else. The correctness requirement states that each party should receive its correct output. Therefore, the adversary must not be able to cause the result of the computation to deviate from the function that the parties had set out to compute. ...  

Haptics (Touch) Update from ACM

 A favorite topic, its part of how we understand reality, with good update here commissioned by ACM:

A Touch of Reality   By Samuel Greengard  Commissioned by CACM Staff

Haptics introduces ways to use virtual reality more effectively, to use robots for a wider array of tasks, and to produce biomechanical replacements for missing limbs that can replicate a sense of feel.

Imbuing robots and other machines with human-like sensory capabilities is an incredibly difficult task. In recent years, machine vision, hearing, and speech have advanced remarkably. Yet, the ability to building machines that can touch and feel the way people do has proven elusive.

All of this is about to change. Haptics—the use of technology to simulate touch, feeling, and motion—is finally hitting its stride. "A confluence of new actuator and sensor technology, along with ever increasing computing power, is pushing the technology into the mainstream," says Jake Rubin, founder and CEO of HaptX, a developer of haptics systems for robots and virtual reality (VR).

The technology introduces ways to use VR far more effectively—particularly for training in industrial settings. Yet haptics also makes it possible to use robots for a wider array of tasks, and to produce biomechanical limbs that replicate a sense of feel. "We are much closer to creating systems that talk to the brain or interact with nerves in ways that seems realistic," says Gregory A. Clark, director of the Center for Neural Interfaces at the University of Utah.

Out of the Body

Simulating human touch is an incredibly complex endeavor. It requires a deep understanding of physiology and neurobiology along with an ability to translate electrical signals into and out of binary code. Although VR, robotics, and biomechanical limbs incorporate and communicate touch in different ways, there's a common denominator, Clark says: a system ultimately must convince the brain that the signals flowing into it from machines are real.

HaptX, for example, has developed highly specialized gloves that use 130 tiny actuators and microfluidic air channels embedded in a fabric. The gloves displace the skin to 2 mm with no latency, producing ultrarealistic sensations; they also offer 6 degrees of freedom. "When you touch a virtual object, the pixel-like bubbles on this material inflate or deflate precisely to produce the same forces over the same area that that object would produce if it touched your hand in real life," Rubin explains.  ... ' 

4 Chatbot Use Cases That Can Help Your Business Drive Revenue

 Another good example of simple delivery of knowledge towards results.

4 Chatbot Use Cases That Can Help Your Business Drive Revenue  By Lydia Roth   in The Alexa Bog

Lydia is Digital Marketing Manager for Alexa. She has 4 years of experience in content marketing and demand generation, and is passionate about helping others grow their marketing know-how.

4 Chatbot Use Cases That Can Help Your Business Drive Revenue

Have you ever interacted with a chatbot that just gets it? A chatbot that’s genuinely interested in helping you find the right solution versus stalking you around a website and asking you if you’re ready to buy something. They understand you as a consumer and give you immediate insight into products or services you’re on the hunt for.

People are not only turning to chatbots for quick answers to product- or service-based questions; they’re actually making purchases from them on the spot. Consider this: 81% of consumers who have made a purchase using a chatbot say they would make another purchase with one.

Having a chatbot can be a nice value add-in for consumers to get more information, and — when executed correctly — they can also help your business make more money.  ... '

Thursday, December 24, 2020

Digital Instruments as Invention Machines

 Regarding 'Knowledge Spillovers', admit I hadn't heard of that, but the term makes quick sense.   Now can you optimize them for invention?   Good article. 

Digital Instruments as Invention Machines

By Pantelis Koutroumpis, Aija Leiponen, Llewellyn D. W. Thomas   Communications of the ACM, January 2021, Vol. 64 No. 1, Pages 70-78


The history of invention is a history of knowledge spillovers. There is persistent evidence of knowledge flowing from one firm, industry, sector or region to another, either by accident or by design, enabling other inventions to be developed.1,6,9,13 For example, Thomas Edison's invention of the "electronic indicator" (US patent 307,031: 1884) spurred the development by John Fleming and Lee De Forest in early 20th century of early vacuum tubes which eventually enabled not just long-distance telecommunication but also early computers (for example, Guarnier10). Edison, in turn, learned from his contemporaries including Frederick Guthrie.11 It appears that little of this mutual learning and knowledge exchange was paid for and can thus be called a "spillover," that is, an unintended flow of valuable knowledge, an example of a positive externality. ...  

The Social Clubhouse

Writeup on 'Clubhouse', mentioned to me in a conversation.    As a possible next big social media thing.   Note the Horowitz support.  As usual,  when I read these things I always end up thinking 'whats the use of that'?    But I have been wrong.

The fatal flaw of Clubhouse   By Mark Schaefer

 Clubhouse appears to be the next hot social media platform.

Backed by the prestigious venture capital firm Andreessen Horowitz and growing faster than Mark Zuckerberg’s bank account, Clubhouse is a platform where users participate in different chat rooms on a wide range of topics. The conversations are audio-only, and when they finish, they disappear forever. It’s like talk radio on steroids — endless, diverse, inclusive. Wading through Clubhouse can be dizzying because there are so many interesting conversations.

Clubhouse is halfway between a podcast and a party. It’s a facilitated discussion, no make-up required. This is why it will displace at least some podcast consumption — you have a chance to actually participate with the hosts, and that’s special.  ... "

Best of the Year Charts

Nathan put out some best of the year charts, on an expected topic.

Best Data Visualization Projects of 2020

It’s been a year. We saw a lot of charts. These are the best ones.

By Nathan Yau

I’ll be honest. I thought about skipping the picks this year. Then I felt that’s no way to finish up 2020. So, here are my favorite visualization projects in no particular order. Of course there was one topic that weighed most heavily, but there were also some other interesting works along the way. ... ' 

California Approves Nuro for Delivery

Have been following Nuro for some time.  With showed
an interesting looking vehicle.  Especially by the examination by Kroger.  An autonomous delivery method. 

Nuro can now charge for robot deliveries in California  By Mitchell Clark   in theVerge 

The permit allows them to charge for human-free deliveries

Nuro is now the first company in California that’s allowed to operate autonomous cars commercially (via TechCrunch). The company received a permit that allowed it to test its robots earlier this year, but this permit will allow the firm to actually charge people for the service.

According to a Medium post by Nuro’s chief legal and policy officer, the company is planning to “announce [its] first deployment in California with an established partner.” Who that partner is remains to be seen, but it’s likely to be a delivery service that can make use of Nuro’s completely driverless Prius vehicles, though the company plans on (literally) rolling out its own custom R2 bots later on.  ... "  

Shoe Fitting Augmented Reality

Likely we will commonly see AR type methods for trying on clothes.

Does the Shoe Fit? Try It On with Augmented Reality

The New York Times   By Kate Conger

Retailers are increasingly reliant on augmented reality (AR) technology to help consumers try on products virtually, hoping to recreate the real-world shopping experience during the current pandemic. Many retailers are using AR offerings from Snapchat parent company Snap, which started adding shopping filters in January, and currently offers AR try-ons for luxury brands. Some companies have developed AR applications exclusively for trying on their own products, like sneaker firm Wanna Kicks; others have teamed with retailers to help create AR experiences focused on their items. Snap in June released a technical library of tools to help developers recognize and classify objects, in order to generate AR filters for Snapchat. Retailers using AR hope virtual try-ons can make product exchanges and returns less likely.  ... " 

Wednesday, December 23, 2020

Worldwide Opinions on Impact of AI

 Results of surveys by country.   Quite bit of detail.  Source is PEW Research. 

Here's how opinions on the impact of artificial intelligence differ around the world

This data on worldwide opinions of artificial intelligence and job automation highlights how education, gender, location, age and income all play a role.

Views on AI are generally more positive among the Asian publics surveyed, with young and educated males the most in favour of automation.

As artificial intelligence (AI) plays a growing role in the everyday lives of people around the world, views on AI’s impact on society are mixed across 20 global publics, according to a recent Pew Research Center survey.

A median of about half (53%) say the development of artificial intelligence, or the use of computer systems designed to imitate human behaviors, has been a good thing for society, while 33% say it has been a bad thing.  ... " 

Machines to Understand Human Goals

Human goals are an important kind of context, so the ability is broadly useful.

Building Machines That Better Understand Human Goals

MIT News   By Rachel Gordon

An algorithm that infers goals and plans, even when those plans might fail, has been developed by Massachusetts Institute of Technology (MIT) researchers. The team used MIT's Gen artificial intelligence (AI) programing platform to integrate symbolic artificial intelligence (AI) planning with Bayesian inference, into the Sequential Inverse Plan Search (SIPS) algorithm. SIPS produced a model that outperformed the baseline Bayesian Inverse Reinforcement Learning model up to 150-fold, and the algorithm inferred goals with 75% accuracy, MIT's Tan Zhi-Xuan said, "This ability to account for mistakes could be crucial for building machines that robustly infer and act in our interests." ... ' 

ORCam Pen Reads Aloud Text

 Seems quite a useful development: 

The OrCam Read is a pen-like AI-based device that reads aloud text to people with reading challenges. OrCam's AI-Based Reading Device Wins CES Innovation Award

The Times of Israel

By Shoshanna Solomon

Israeli artificial intelligence (AI) device developer OrCam Technologies has been named to receive the CES 2021 Innovation Awards Best of Innovation award in the "accessibility" category for its OrCam Read digital reader, a pen-sized assistive handheld device developed for people with reading challenges. The device is designed to read aloud full pages or screens of text from any printed or digital surface, without an Internet link. The device's AI algorithms enable intuitive point-and-click function, triggering two precision laser-guidance options to read highlighted text or target a point to start reading. Its text-to-speech reading engine also has an interactive feature integrating computer vision and natural language understanding as a voice-activated assistant. ... " 

AI Predictions

Some end of year predictions for next year.   I much like the title mentions, which I do think are important.  But don't necessarily fully agree with all of the futures mentioned.   Still includes the opinions of many practitioners, which is useful.

2021 Predictions: ML Automation, K-anonymity, Hyperautomation, Semantic Graphs  in AITrends

[Ed. Note: We have heard from a range of AI practitioners for their predictions on AI Trends in 2021. Here are predictions from a selection of those writing.] 

Florian Douetteau, CEO and co-founder of Dataiku

Inclusive engineering will begin to make its way into the mainstream to support diversity. In order to ensure diversity is baked into their AI plans, companies must also commit the time and resources to practice inclusive engineering. This includes, but certainly isn’t limited to, doing whatever it takes to collect and use diverse datasets. This will help companies to create an experience that welcomes more people to the field — looking at everything from education to hiring practices.    ... " 

There will be more of an organizational commitment to putting humans and diversity at the center of AI development. Companies will look to include people who are representative of those who will use the algorithms if they want to truly reduce bias and foster diversity. While most training datasets have been developed against a small percentage of the population, companies will now look to consider expanding their scope to design training datasets that are all-inclusive. The more inclusive the group building the AI and the datasets, the less the risk for bias.  ... " 

Google: Duo and Meet: Making Moments that Matter

In the Google Blog, thoughts on making meetings more productive, and their package 'Meet'.   Seems I have used most other packages lately , but not their Meet.  Now connected with Gmail.  Now free and unlimited through the holidays.  Though it seems they have lost some of the inertia caused by the pandemic.    But I like Google's thoughtful look at the problem of how do you make these connections address meaningful goals, whether they are connected to home or business.

Google Meet and Duo help you share moments that matter   By Dave Citron, Director of Product Management, Google Duo & Google Meet

Without a doubt, 2020 was the year of video calling. And for us, that meant making sure every student, team, and family could jump on a call from any device and have a reliable, safe experience. Google Duo and Google Meet hosted over one trillion minutes of video calls globally. For perspective, that’s equal to more than 18 billion hour-long virtual workouts in a single year! 

Here’s a recap of what we’ve done so far.

Productive and engaging meetings at home, work, and school 

In 2020, Meet was put to the test. Our team had to really think through how virtual meetings could bring the key part of what in-person meetings provide: human connection. We launched new features like 49-tile layout, noise cancellation, background blur, and low-light mode plus live captions in five languages to help everyone follow along on the call. 

Earlier this year, we made Meet free for everyone. We also announced that with your Gmail account, Meet calls are unlimited through March 31, 2021 so that families can enjoy their holiday traditions without interruptions. Speaking of Gmail, we added a Meet tab in Gmail, so that with one tap, people can jump from an email to a video call. We also brought Meet to Nest Hub Max and Chromecast to help people get up, move around, and have hands-free calls at home. 

In large group settings like team all-hands or a classroom, it gets harder for people to speak up and to engage everyone on the call. With Meet, participants can use Q&A and hand raising tools, polls and breakout rooms. Organizations and moderators have more control too, keeping their meetings and participants safe, including advanced anti-abuse features, that allow for an enjoyable, safe experience for all. And in 2021, as many companies evaluate a flexible working model, we’ve designed Meet to work with our Series One hardware kits, created to deliver inclusive audio and video clarity that makes you feel like you're all together. So whether you're a Google Workspace subscriber relying on Meet's enterprise-grade functionality, or using Meet’s free version to safely connect with others near and far, Meet has you covered. ... ' 

Chinese Quantum Computing

Quantum computing is expected to be able to be able to 'break' some forms of encryption,  thus the concerns mentioned.

China joins Google in claiming quantum supremacy with new technology, ratcheting up RSA decryption concerns.

China’s top quantum-computer researchers have reported that they have achieved quantum supremacy, i.e., the ability to perform tasks a traditional supercomputer cannot. And while it’s a thrilling development, the inevitable rise of quantum computing means security teams are one step closer to facing a threat more formidable than anything before.

Researchers from the University of Science and Technology of China explained in the journal Science they were able to get a system they named Jiuzhang to perform a calculation in minutes that would have taken a traditional supercomputer an estimated 10,000 years to solve.

The team joins Google, which claimed it achieved quantum supremacy in Oct. 2019 using a “supercold, superconducting metal,” according to WIRED. IBM has also entered the quantum computing fray, while leveling criticism against Google’s claims of supremacy.

Now, the Chinese researchers have claimed quantum supremacy using a quantum computation called Gaussian boson sampling (GBS), their paper explained, which uses particles of light sent through an optical circuit, measuring the output. This means there are now multiple proven quantum-computing technologies, with surely more to come.

The security concern is that quantum computers will be able to crack RSA public key cryptography, used to protect data in transit. That means security teams will have to pivot to new post-quantum cryptography solutions. A conservative estimate from a 2019 DigiCert report said teams will need to have protections from quantum computing breaches in place by 2022.   ... " 

Talking Microservices

 Had cause to talk the concept recently,  in InfoQ, here is a somewhat technical look at the concept and its uses and details. Most of the details and useful links at the clickthrough.

Reviewing the Microservices Architecture: Impacts, Operational Complexity, and Alternatives  by Wesley Reisz, Yan Cui in InfoQ

Key Takeaways

Although we can find case studies about microservices migrations, there are still a lot of companies in the industry that haven’t touched the microservices strategy yet.

The current microservices approaches are more complex than they used to be. We build more complex systems, with a more complex architecture, so we have a more complex landscape and a deeper learning curve as consequence.

Monitoring and tracing microservices are some of the biggest challenges, besides its complexity.

The event-driven architecture is a great way to build microservices, especially in terms of communication between different services.

Wes Reisz moderated an InfoQ Live roundtable on the impact of microservices, dealing with operational complexity, and alternatives to the microservices model. The participants were Leif Beaton (NGINX senior solutions architect), Yan Cui (independent AWS and serverless consultant), and Nicky Wrightson (Skyscanner principal engineer).

Peter Rodgers first introduced the term "micro-web-services" during a presentation at the 2005 Web Services Edge Conference, going against the conventional thinking at the time. The SOAP service-oriented architecture peaked in 2005 and Rodgers was arguing for RESTful services. He described in his presentation how a well-designed micro-web-services platform combines the underlying architectural principles of the web and REST services in Unix-like scheduling and pipelines to provide flexibility and simplicity in service-oriented architectures.

After six years of innovation and thought about this problem, a May 2011 workshop for software architects coined the term "microservice" to describe what participants were seeing and employing. By the next spring, the community embraced "microservices" to describe this architectural style.   ... '

Tuesday, December 22, 2020

More on SolarWind: A Kill Switch

 FireEye on the SolarWind backdoor, from last week.

And work by Microsoft on the problem:

FireEye, Microsoft create kill switch for SolarWinds backdoor  By Lawrence Abrams in BleepingComputer

Microsoft, FireEye, and GoDaddy have collaborated to create a kill switch for the SolarWinds Sunburst backdoor that forces the malware to terminate itself.

This past weekend it was revealed that Russian state-sponsored hackers breached SolarWinds and added malicious code to a Windows DLL file used by their Orion IT monitoring platform.

This malicious DLL is a backdoor tracked as Solarigate (Microsoft) or Sunburst (FireEye) and was distributed via SolarWinds' auto-update mechanism to approximately 18,000 customers, including the U.S. Treasury, US NTIA, and the U.S. Department of Homeland Security.  

As part of a coordinated disclosure with Microsoft and SolarWinds, FireEye released a report on Sunday with an analysis of the supply chain attack and how the Sunburst backdoor operates.  This research revealed that the Sunburst backdoor would connect to a command and control (C2) server at a subdomain of avsvmcloud[.]com to receive 'jobs', or commands to execute. ... " 

(as the article suggests, the malware may have created other backdoors, so this will not necessarily remove the problem) 

Hyundai Buys Boston Dynamics

More on this, implications?  From what I have seen the largely remotely operated tele-robotic systems have particular value in remote security and remote data and information gathering and analysis.   Adding a particularly impactful remote presence.  Much more in the article below:

Hyundai Buys Boston Dynamics for Nearly $1 Billion. Now What?    in IEEE Spectrum By Evan Ackerman

Boston Dynamics contemplates its future as it gets swallowed (again) by a giant corporation  

Boston Dynamics, once owned by Google's parent company, Alphabet, and then by Japanese conglomerate SoftBank, now belongs to South Korean carmaker Hyundai.  .... '

Amazon Wants to Train 29 Million People to Work in the Cloud

 We are already there and working hard, but how that will be done efficiently still remains to be established.  

Amazon Wants to Train 29 Million People to Work in the Cloud

The Wall Street Journal,  Chip Cutter

Amazon plans to help train 29 million people worldwide to work in the cloud by 2025, by building on existing programs and adding new ones in collaboration with nonprofits, schools, and others. The online retailer said the goal is to equip people with the skills for working in cloud computing at client companies aiming to fill high-technology positions. Amazon added that the free training could help people preparing for entry-level support positions, or expand existing engineers' expertise in areas like machine learning or cybersecurity. Cloud computing has become a core capability for many companies as they accelerate their adoption of digital tools. Amazon Web Services' Teresa Carlson said, "We need our customers to have the right skills if they're going to go through a digital transformation. ... 

Toward Hybrid Approaches of Problem Solving

 We knew this long ago, but only since approaches like machine learning have done some surprising things have we become ready for more contextually useful advances.  

Ushering in the third wave of AI   By Tolga Kurtoglu Head of Global research at Xerox,  in TechRadar

Breaking down the language barrier between AI and humans

Today, artificial Intelligence (AI) helps you shop, provides suggestions on what music to listen to and what shows to watch, connects you with friends on social media and even drives your car.

As more companies focus their efforts on AI-based solutions, 2020 is shaping up to be a turning point as we begin to witness the third wave of AI — when AI systems not only not learn and reason as they encounter new tasks and situations, but have the ability to explain their decision making.

Where We Are Now

The first wave of AI focused on enabling reasoning over narrowly defined problems, but lacked any learning capability and poorly handled uncertainty. Financial products like Turbotax and Quickbooks, for example, are able to take information from a situation where rules have previously been defined and work through it to achieve a desired outcome. However, they are unable to operate beyond the previously defined rules.

The second wave, which we are in the midst of right now, is AI that has nuanced classification and prediction capabilities, but no contextual capability and minimal reasoning capability. Major machine learning-based AI platforms like IBM’s Watson and Salesforce’s Einstein are good examples as they are able to synthesize large amounts of data to provide insight and answers, but are not able to truly understand or explain how they got to that answer.

The Third Wave: AI That Understands and Reasons in Context

So how do we progress from the second wave to the third wave?

The starting point is to make AI explainable, more transparent. AI algorithms will continue to increasingly interact with humans across many industries — in our homes, our cars, and our clothing. If they continue to evolve the way they have been, and they are black boxes in nature, there is a potential concern with transparency and eventually with trust. To avoid these concerns, these systems should be transparent in a way that they can explain their work, including the assumptions they made, the different options they considered and eventually why they came up with the answer that they provided.

Right now, there is an opportunity to establish trust between the system and the human. Explainability and transparency are a starting point to something much broader—the collaboration between humans and computers to solve the most complex problems in the world.  ... " 

Limitations of Deep Learning

About limitations of deep learning problems with higher dimensional data. Technical.

Geometric Deep Learning Advances Data Science   By Samuel Greengard

Communications of the ACM, January 2021, Vol. 64 No. 1, Pages 13-15


Deep learning has transformed numerous fields. In tackling complex tasks such as speech recognition, computer vision, predictive analytics, and even medical diagnostics, these systems consistently achieve—and even exceed—human-level performance. Yet deep learning, an umbrella term for machine learning systems based primarily on artificial neural networks, is not without its limitations. As data becomes non-planar and more complex, the ability of the machine to identify patterns declines markedly.

At the heart of the issue are the basic mechanics of deep learning frameworks. "With just two layers, a simple perceptron-type network can approximate any smooth function to any desired accuracy, a property called 'universal approximation'," points out Michael Bronstein, a professor in the Department of Computing at Imperial College London in the U.K. "Yet, multilayer perceptrons display very weak inductive bias, in the sense that they assume very little about the structure of the problem at hand and fail miserably if applied to high-dimensional data."  .... ' 

Monday, December 21, 2020

Is it the Year the Password will Die?

Still a few things to think through,  but looking forward to it. 

Microsoft: 2021 is the Year Passwords Die    By Neowin

Removing the password. 

Microsoft is making an effort to transition people to passwordless solutions.

Microsoft has been a proponent of passwordless technology for quite some time, saying that it wants traditional and unsafe passwords to die. To that end, it has invested in various solutions over the past few years such as Windows Hello, Microsoft Authenticator, FIDO2 security keys, and a palm vein authentication system, among other things.

Now, the company has highlighted the strides it made to kill off passwords in 2020, and has stated that it plans to make them a thing of the past for all its customers in 2021.

Microsoft noted that almost 80% of all cyberattacks target passwords, and one in 250 corporate accounts get compromised each month due to this. That said, the company is making an effort to transition people to passwordless solutions. In November 2019, 100 million people were using Microsoft's passwordless sign-in. This number grew to 150 million by May 2020, which goes to show how millions of people are ready to ditch passwords due to the inconvenience of remembering them, coupled with how insecure they can be.

From Neowin

Thinking Assistive Disaster AI

A clear example of the interactions of sensors, logic, decisions and leveraging experience.  Some general ideas of the possibilities.

Assistive AI in Decreasing Natural Disaster Damage

Assistive AI is an invaluable tool to minimize the damages of natural disasters.  By Jack Williams  in SmartDataCollective

 A few years ago, we talked about the benefits of using AI and big data in disaster relief. We have recently seen how important these applications are, as the damage of natural disasters is worse than ever.

Mother Nature has been busier than ever this year, with natural disasters adding to the stress of the COVID-19 pandemic. A number of hurricanes have wreaked havoc on the Gulf Coast, and wildfires recently devastated the West Coast. Unfortunately, these severe weather events now occur more frequently. In seven of the last 10 years, there have been 10 or more weather-related disasters with costs exceeding $1 billion each. Among the 115 costly catastrophes since 2010 were so-called “1,000-year floods,” Category 5 hurricanes, and the most massive wildfires in history.

These problems put the importance of big data and AI to the test, as we strive to fight the problems they cause.

Worsening Disasters Highlight the Need for New Assistive AI Solutions

The problem is that, despite their calamitous effects, many natural disasters can be surprisingly difficult to spot in their beginning stages. For example, the 2018 Camp Fire in California that tore through large portions of the state, eventually burning over 150,000 acres and destroying more than 18,000 structures, began in the blink of an eye; it spread over the first seven miles in less than 90 minutes.  ... " 

Security with Threat Hunting

Is this ultimately the means to thwarting security failures?

Threat Hunting Offsets the Technology Gaps

Caitlin Mattingly, Recorded Future

Our guest this week is John Ayers, Executive Vice President, Chief Strategy Product Officer and head of Security Operations at Nuspire, a managed security services company.

Our conversation centers on John’s assertion that threat hunting has become an indispensable element of security strategy for many organizations. He explains the evolution of threats that led him to that conclusion, and we’ll discuss how organizations can best approach implementing threat hunting into their own defensive plans.

This podcast was produced in partnership with the CyberWire.  ...   

Why Business Transformation Projects Often Fail

 Thoughts and links from an IBM exec we worked with: 

Why Business Transformation Efforts Often Fail  by Irving Wladawsky-Berger

A 2019 study found that the average digitization level across all industry sectors was only around 25% of their ultimate potential.  But, the pandemic has now made the case for accelerating the digital transformations firms and economies were forced to make to help them cope with the crisis.  But, how successful are such major transformation likely to be?

A few months ago I received an email from the Harvard Business Review recommending a classic article from its 50 Best Selling collection that could shed light on this important and timely question, - Leading Change: Why Transformation Efforts Fail, originally published in 1995 by (now emeritus) Harvard professor John Kotter.

“Over the past decade, I have watched more than 100 companies try to remake themselves into significantly better competitors,” wrote Kotter.  These included large and small organizations, US and non-US based, some that were on their knees and some that were doing quite well.  “But, in almost every case, the basic goal has been the same: to make fundamental changes in how business is conducted in order to help cope with a new, more challenging market environment.”  ... ' 

Eye Movements Predicting Mental Health

What appears to be a reasonable pattern detection type application

AI gauges the mental health of cancer patients through eye movements

It's an early project, but shows potential.

Jon Fingas, @jonfingas  in Engadget

Good mental health is important, and early research suggests AI might help diagnose that health for people who are under a particularly heavy strain. Scientists have developed a combination of deep learning algorithms that use eye tracking to gauge the mental health of cancer patients after surgery. Ideally, this would help spot patients likely to be suffering from anxiety or depression when a human can’t perform an initial psychological assessment.

The system uses a mix of a convolutional neural network and long short-term memory algorithms to study the eye movements of patients wearing tracking glasses (in this case, Tobii Pro 2 glasses) while they contemplate artwork. The AI used the gaze and pupil position data from those glasses to determine how likely someone was to raise concerns on established hope, anxiety and mental wellbeing questionnaires they would fill out later.  ... ' 

Netflix Digital Experiments with Content Decisions

With lots of interesting links to presentations I am looking at.   How Netflix works with its content has always interested me.   Exploring this:

Netflix at MIT CODE 2020

Netflix Technology Blog

In November, Netflix was a proud sponsor of the 2020 Conference on Digital Experimentation (CODE), hosted by the MIT Initiative on the Digital Economy. As well as providing sponsorship, Netflix data scientists were active participants, with three contributions.

Eskil Forsell and colleagues presented a poster describing Success stories from a democratized experimentation platform. Over the last few years, we’ve been Reimagining Experimentation Analysis at Netflix with an open platform that supports contributions of metrics, methods and visualizations. 

This poster, reproduced below, highlights some of the success stories we are now seeing, as data scientists across Netflix partner with our platform team to broaden the suite of methodologies we can support at scale. Ultimately, these successes support confident decision making from our experiments, and help Netflix deliver more joy to our members!   ... ' 

Sunday, December 20, 2020

Decoding the Neural Activity of Computer Programming

Intriguing, especially as it differs from other intense activity.    Is age dependence very different from individual dependence?  Language vs logic?  Kinds and method of scans? 

Scientists Decode Neural Mechanisms of Computer Programming  By News-Medical Life Sciences

Johns Hopkins University scientists mapped the brain activity of expert computer programmers while they puzzled over code, to find the neural mechanics behind this increasingly vital skill.

Johns Hopkins University (JHU) scientists mapped computer programmers' brain activity during coding exercises to uncover the underlying neural processes.

JHU's Yun-Fei Liu scanned the brain activity of 15 experienced programmers skilled in Python via functional magnetic resonance imaging, as they worked on coding questions.

The exercise activated the brain region responsible for logical reasoning in each participant, with coding largely skewing to the left hemisphere, which correlates with language.

The JHU researchers hope to determine if programming, like learning language, is easier for young people.

JHU's Marina Bedny said, "It could be that our education system is wrong, and we should be teaching kids to code in middle school or else they're missing an opportunity to be the best they can."

From: News-Medical Life Sciences

Occams Razor and Ensemble Learning

Thoughtful piece.  Much more at the link.   Here the intro:

Ensemble Learning Algorithm Complexity and Occam’s Razor   by Jason Brownlee on December 21, 2020 in Ensemble Learning 

Occam’s razor suggests that in machine learning, we should prefer simpler models with fewer coefficients over complex models like ensembles.

Taken at face value, the razor is a heuristic that suggests more complex hypotheses make more assumptions that, in turn, will make them too narrow and not generalize well. In machine learning, it suggests complex models like ensembles will overfit the training dataset and perform poorly on new data.

In practice, ensembles are almost universally the type of model chosen on projects where predictive skill is the most important consideration. Further, empirical results show a continued reduction in generalization error as the complexity of an ensemble learning model is incrementally increased. These findings are at odds with the Occam’s razor principle taken at face value.

In this tutorial, you will discover how to reconcile Occam’s Razor with ensemble machine learning.

After completing this tutorial, you will know:

heuristic that suggests choosing simpler machine learning models as they are expected to generalize better. The heuristic can be divided into two razors, one of which is true and remains a useful tool and the other that is false and should be abandoned.

Ensemble learning algorithms like boosting provide a specific case of how the second razor fails and added complexity can result in lower generalization error.

Let’s get started.  ...  "

Identification of Plants and Weeds

Have been testing a plant recognition system for some time.   Effectiveness varies, but continues to improve.

Deere's Farm Version of Facial Recognition Coming to Fields in 2021   By CNBC

Agriculture giant Deere & Co. plans to roll out a system next summer that combines machine vision and machine learning to improve the identification of individual plants and weeds.

Deere's Jahmy Hindman said neural network models could be trained to only spray weeds in crop fields, killing everything except genetically modified plants designed to survive chemical applications.

Said Hindman, "We are interested in being able to manage each plant over the course of its life, minimizing inputs and maximizing productivity."

The technology would take pictures of plants, and a machine cruising the field would make the decision to spray in just seconds.

Jeffries' Stephen Volkmann said, "See-and-spray is one of several advanced farming technologies that seem to be moving closer to an inflection point," but commercialization of full plant recognition technology is still a few years away. ... '

Effectively Mapping Business Terms

 Ultimately what can make such approaches work.   Have worked with TopQuadrant in the past.

WEBINAR: Automating the Mapping of Data Elements to Business Terms

Join us for this popular Webinar, offered again at more convenient time for our European and beyond audience! January 7, 2021 at 8:00 AM ET.

A business glossary is a collection of terms important to a business. It can capture valid definitions of values and business rules for data, sharing this key knowledge across the organization. A business glossary can be used to link IT assets such as data elements to terms that describe the data; thereby, connecting business and IT and enhancing organizational collaboration and productivity.

A well-defined and operationalized business glossary saves time for employees and increases consistency, standardization and data integrity for the entire enterprise. However, the process for connecting data sources to business terms presents a challenge for many organizations. With hundreds of data sources and many thousands of data elements, it is a large undertaking.

In this webinar, we will describe how TopBraid EDG can help you automate this process, including: 

Using Data Element Names to Map to Business Terms

Using Data to Map to Business Terms 

 More details about the webinar and speakers.   and Registering.

After registering, you will receive a confirmation email containing information about joining the webinar. 

Amazon Group Calling and Captioning

 Had just heard of this, ways to make group calls more efficient is very useful.  About to try.

Amazon’s Echo devices gain group-calling and call captioning features  By Kelly Hodgkins in DigitalTrends

Can’t get together with your family this year for Christmas? You are not alone. Instead of gathering for a large holiday celebration, many people are opting to meet virtually. To help people stay connected while apart, Amazon added new audio and video calling features to Alexa and its Echo devices starting today.

With Amazon’s group-calling feature, people can connect to seven friends and family members on a single hands-free video or audio call. All it takes is a single person to create and name a group using Alexa. Once the group is formed, other friends and family members can join by telling Alexa to “Call my (Group Name).” This group-calling feature is available on compatible Echo devices, including the Echo, Echo Dot, and Echo Show.  

Amazon also added a new closed captioning feature that is available in both audio and video calls. During the call, Amazon will convert the other person’s voice to text in near real time. Captioning works with U.S. English and is available on all Echo Show devices.  ... ' 

AI For Vaccine Rollout and Tracking

In particular for tracking deployment in complex supply chains with complex data. 

How AI is Helping with COVID-19 Vaccine Rollout and Tracking 

December 17 

Covid-19 Coronavirus Vaccine vials in a row macro close up in AITrends  By John P. Desmond, AI Trends Editor  

AI has been employed since the early days of the COVID-19 pandemic to track the spread of positive cases, to crunch through thousands of scientific papers to search for treatment options and to help develop a vaccine. Now AI and other digital tools are being deployed to manage complex supply chains for the vaccine. 

With the third highest number of coronavirus cases in Europe after France and Italy, according to data from Johns Hopkins University, the UK is the first country to distribute the Pfizer vaccine. The UK has 1.7 million confirmed cases in a population of close to 68 million people, Tracking side effects from the vaccine rollout is a huge task, UK health officials have said, according to an account from Nasdaq. To help meet the challenge, the UK Medicines & Healthcare products Regulatory Agency (MHRA) recently partnered with the UK unit of Genpact, the global professional services firm specializing in digital transformation. The company is integrating components of its AI software suite with the government’s website where adverse effects are reported. 

“When a vaccine gets distributed at scale and speed, a technology solution needs to track the batch and lot numbers to know exactly where each dose is and who received it,” stated Eric Sandor, drug safety AI lead at Genpact. “There’s a lot of information, in a number of different formats, and it’s very manually intensive to try to codify it in a way that makes sense. AI will help with processing all that data faster than humans can. It’s quite complicated at scale, but is a critical element to overall public health.”    ... ' 

IBM and Salesforce Collaborate

New efforts underway to address verification of health status.

IBM and Salesforce Join Forces to Help Organizations and Individuals Verify Vaccine and Health Status

ARMONK, N.Y. and SAN FRANCISCO, Dec. 18, 2020 /PRNewswire/ -- IBM (NYSE: IBM) and Salesforce (NYSE: CRM) today announced that they are partnering to help organizations as they strive to safely reopen public places and provide individuals with a verifiable and privacy-preserving way to manage and share their vaccination and health status in the wake of the COVID-19 pandemic. As part of this offering, IBM Digital Health Pass will integrate with the Salesforce Work.com platform. Integration of the  ... "

Saturday, December 19, 2020

Electric Amoeba efficiently Solves TSP

Classic problem, usefully solved?

Electronic Amoeba' Finds Approximate Solution to Traveling Salesman Problem in Linear Time  By Hokkaido University (Japan)

Researchers in Japan developed electronic amoeba, an analog computer inspired by a single-celled amoeboid organism, to solve the traveling salesman problem.

Researchers at Hokkaido University and Amoeba Energy in Japan have, inspired by the efficient foraging behavior of a single-celled amoeba, developed an analog computer for finding a reliable and swift solution to the traveling salesman problem — a representative combinatorial optimization problem.

Many real-world application tasks such as planning and scheduling in logistics and automation are mathematically formulated as combinatorial optimization problems. Conventional digital computers, including supercomputers, are inadequate to solve these complex problems in practically permissible time as the number of candidate solutions they need to evaluate increases exponentially with the problem size — also known as combinatorial explosion. Thus new computers called "Ising machines," including "quantum annealers," have been actively developed in recent years. These machines, however, require complicated pre-processing to convert each task to the form they can handle and have a risk of presenting illegal solutions that do not meet some constraints and requests, resulting in major obstacles to the practical applications.

These obstacles can be avoided using the newly developed "electronic amoeba," an analog computer inspired by a single-celled amoeboid organism. The amoeba is known to maximize nutrient acquisition efficiently by deforming its body. It has shown to find an approximate solution to the traveling salesman problem (TSP), i.e., given a map of a certain number of cities, the problem is to find the shortest route for visiting each city exactly once and returning to the starting city. This finding inspired Professor Seiya Kasai at Hokkaido University to mimic the dynamics of the amoeba electronically using an analog circuit, as described in the journal Scientific Reports. "The amoeba core searches for a solution under the electronic environment where resistance values at intersections of crossbars represent constraints and requests of the TSP," says Kasai. Using the crossbars, the city layout can be easily altered by updating the resistance values without complicated pre-processing.

From Hokkaido University (Japan)

Quantum Computing Real Problems

 More advances in the quantum space.

Tiny Quantum Computer Solves Real Optimization Problem  By Chalmers University of Technology (Sweden)

December 18, 2020

Researchers at Sweden's Chalmers University of Technology showed that a small but well-functioning quantum computer was able to solve a small portion of a real logistics problem in the aviation industry.

Quantum computers already have shown their ability to surpass ordinary computers in solving certain tasks, but until now, those tasks have had no practical use.

The researchers successfully executed the Quantum Approximate Optimization Algorithm (QAOA) on a quantum computer comprised of a processor with two qubits to solve the problem of assigning aircraft to routes in practice.

Although their demonstration involved just two airplanes, they simulated solving the same optimization problem for up to 278 aircraft, which would require 25 qubits.  ... 

Chalmers' Giulia Ferrini said, "The results remained good as we scaled up. This suggests that the QAOA algorithm has the potential to solve this type of problem at even larger scales."

From Chalmers University of Technology (Sweden)

On Predictability of Elections

Well done with useful links

On Predictability of Elections

By Irving Wladawsky-Berger

A collection of observations, news and resources on the changing nature of innovation, technology, leadership, and other subjects.

Are There Limits to the Predictability of Elections?   

Friday, December 18, 2020

SANS Institute: SolarWinds Attack: What You Need to Know

Presentation:  The SolarWinds Supply Chain Attack:  What you Need to Know:  https://www.youtube.com/watch?v=qP3LQNsjKWw

SANS Emergency Webcast: What you need to know about the SolarWinds Supply-Chain Attack

45,435 views•Dec 14, 2020

SANS Institute

On Dec 13, 2020, Solarwinds, an IT company that creates software for network management, stated they were investigating an incident that appears to be the product of a “highly-sophisticated, targeted and manual supply chain attack by a nation-state.” SolarWinds said they are in contact with the FBI and that a vulnerability which existed until the March-June 2020 timeframe was leveraged to take advantage of their Orion software product.

The attack is a supply-chain based attack in which the adversary can leverage the software’s update mechanism.  The Solarwinds attack has been linked to the Treasury Department and FireEye compromises at this time.

Information is being released continuously by those investigating the incidents across the thousands of organizations that use SolarWinds, including governments, militaries, and commercial entities around the world.  As indicators of compromise continue to be released, organizations and their incident response teams should prioritize hunting for adversary behaviors and Tools, Techniques, and Procedures (TTPs) associated with their SolarWinds installs, as that platform could be leveraged as a launching point into their organization.

Participants will learn about:

- The latest information regarding the Solarwind's incident and the mechanics of the supply chain attack.

- Any known detection mechanisms, including IOCs, have been released at this point.

- How the incident could impact organizations that use SolarWinds and where to begin investigations.

Speaker Bio

Jake Williams @malwarejake is a SANS analyst, senior SANS instructor and course author. Jake spent more than a decade in information security roles at several government agencies, developing specialties in offensive forensics, malware development and digital counterespionage. Jake is the founder of Rendition Infosec, which provides penetration testing, digital forensics and incident response, expertise in cloud data exfiltration, and the tools and guidance to secure client data against sophisticated, persistent attacks on-premises and in the cloud.

SANS is the most trusted and by far the largest source for information security training and security certification in the world. It also develops, maintains, and makes available at no cost, the largest collection of research documents about various aspects of information security, and it operates the Internet's early warning system - the Internet Storm Center. .... 

SpaceX vs Kuiper on Internet Delivery

Been following SpaceX for some time, and is use of multiple satellites to fill remote areas with high speed internet service.   Now in limited test.   And considering Amazon's competitive Kuiper, here is an outline of their current efforts.  How will this effect the current land based cable services that fill in internet services? 

Amazon’s Satellite Internet Antenna Pulls 400Mbps During Testing   By Ryan Whitwam in ExtremeTech

For years, satellite internet has been a last resort for people living in remote areas. SpaceX has made waves offering its new Starlink service to select regions with much higher speeds than traditional satellite connections — testers are reporting as much as 150Mbps down. Amazon thinks it can do better with its new antenna technology. The company says its prototype for satellite internet is managing 400 Mbps right now, but we don’t know when consumers will be able to put it to the test. 

You’ve probably heard from people stuck with satellite internet from companies like HughesNet and Viasat. Most likely, they were not happy with the multi-second latency and anemic speeds, but it’s better than nothing. That’s actually what SpaceX called its Starlink beta this year: Better Than Nothing. Starlink is currently limited to the northern US and southern Canada, and it requires a hefty $500 setup fee. That includes the company’s satellite dish, which connects to the network over the Ka wireless band.

Amazon’s upcoming Project Kuiper service will be similar to Starlink, but the company claims its prototype Ka phased-array antenna will give it the edge. Amazon’s goal is to reduce the size and cost of the hardware — after all, every customer will need one to access the Kuiper network. However, that’s difficult with Ka-band equipment, which needs more physical separation between transmission (27-30GHz) and reception (17-20GHz) hardware due to the wide range frequencies.   ... " 

Find the Warning Signals

A classic kind of problem we addressed in practice: can we find signals of, or predictive signals of future occurrence?   Powerful idea. How possible is this, very generally in streams of data?  Connection to risk analysis?

Method Finds Hidden Warning Signals in Measurements Collected Over Time

MIT News  Daniel Ackerman

A new technique developed by researchers at the Massachusetts Institute of Technology (MIT) and Spain's Rey Juan Carlos University uses a generative adversarial network (GAN) to identify anomalies in time series data. The TadGAN approach could be used to detect and respond to significant changes in a range of high-value systems. The researchers aimed to create a general framework for anomaly detection that could be applied across industries. TadGAN can distinguish between normal and anomalous data points by checking for discrepancies between the real-time series and a fake GAN-generated time series. Supplementing the GAN with an autoencoder algorithm helped the researchers strike a balance between being vigilant and raising too many false alarms. In anomaly detection tests on 11 datasets, TadGAN outperformed ARIMA, the traditional approach to time-series forecasting, for eight datasets. Said MIT's Sarah Alnegheimish, "We want to mitigate the stigma about artificial intelligence not being reproducible."

Using AI to Control a Military Jet

As described a considerable application, combining  autonomous signals and human decisions.

In a First, USAF Uses AI on Military Jet

The Washington Post  By Aaron Gregg

U.S. Air Force (USAF) officials said artificial intelligence (AI) was used for the first time on a military aircraft, when an algorithm called ARTUµ on Tuesday controlled radar and tactical navigation on a U-2 Dragon Lady spy plane during a training flight. USAF’s Josh Benedetti said ARTUµ intentionally lacked a manual override to "provoke thought and learning in the test environment." People involved in the test said the AI performed specific tasks while separated from the flight controls, which a human pilot handled. Assistant Air Force Secretary Will Roper said ARTUµ was trained against an opposing computer to scan for oncoming missile and missile launchers, and had the final say on where to orient the plane's sensors. ARTUµ is based on open source algorithms, and Roper said the purpose of test is "to shock the Air Force and the [Defense] Department as a whole into how seriously we need to treat AI teaming."

Solarwinds Intrusion Hack is Broadly Felt

Continued attacks and potential consequences, as indicated by Solarwinds intrusion.  

More Hacking Attacks Found as Officials Warn of ‘Grave Risk’ to U.S. Government  By The New York Times

The scope of the hacking appears to extend beyond nuclear laboratories and Pentagon, Treasury and Commerce Department systems.

 Federal officials issued an urgent warning on Thursday that hackers who American intelligence agencies believed were working for the Kremlin used a far wider variety of tools than previously known to penetrate government systems, and said that the cyberoffensive was "a grave risk to the federal government."  ... 

The discovery suggests that the scope of the hacking, which appears to extend beyond nuclear laboratories and Pentagon, Treasury and Commerce Department systems, complicates the challenge for federal investigators as they try to assess the damage and understand what had been stolen. 

"A good defense isn't enough; we need to disrupt and deter our adversaries from undertaking significant cyberattacks in the first place," Mr. Biden said, adding, "I will not stand idly by in the face of cyberassaults on our nation."... " 

Thursday, December 17, 2020

Robots Encourage Risk-Taking Behavior in Humans

 Fascinating premise, but highly context dependent,  would not go into a project with this idea strongly held.  Has some relationship to the 'Media Equation' concept, where trust where we trust robots more because they apparently have no hidden motives.    We used that method to drive us to assistant models. 

'The Robot Made Me Do It': Robots Encourage Risk-Taking Behavior in Humans By University of Southampton

Robots can encourage humans to take greater risks in a simulated gambling scenario than they would if the robot is silent, research shows. Increasing understanding of whether robots can affect risk-taking could have clear ethical, practical, and policy implications, according to "The Robot Made Me Do It: Human-Robot Interaction and Risk-Taking Behavior," published in Cyberpsychology, Behavior, and Social Networking.

The research involved 180 undergraduate students taking the Balloon Analogue Risk Task (BART), a computer assessment that asks participants to press the spacebar on a keyboard to inflate a balloon displayed on the screen. With each press of the spacebar, the balloon inflates slightly, and 1 penny is added to the player's "temporary money bank." The balloons can explode randomly, meaning the player loses any money they have won for that balloon and they have the option to "cash-in" before this happens and move on to the next balloon.

One-third of the participants took the test in a room on their own, one third took the test alongside a robot that only provided instructions but was silent the rest of the time, and the third group took the test with the robot providing instruction as well as speaking statements such as, "Why did you stop pumping?"

The results showed that the group who were encouraged by the robot took more risks, blowing up their balloons significantly more frequently than those in the other groups did. They also earned more money overall.

"On the one hand, our results might raise alarms about the prospect of robots causing harm by increasing risky behavior," says Yaniv Hanoch, associate professor in risk management at the University of Southampton who led the study. "On the other hand, our data points to the possibility of using robots, and AI, in preventive programs such as anti-smoking campaigns in schools, and with hard to reach populations, such as addicts."

From University of Southampton