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

Friday, September 22, 2017

Delivering Perishables to the Smart Home

We tested a number of solutions to perishable delivery in out smart home labs. Obviously this requires some expensive refitting of the home.

Walmart tests new delivery drop-off point — the customer’s fridge
By Deena M. Amato-McCoy

Walmart’s new grocery delivery program could give it a huge edge in the online ordering game.

The discount giant is testing a concept that will not only deliver fresh groceries, but also enable a delivery person to enter customers’ homes and put away perishables in their refrigerator. Walmart, which announced the news in a blog on its website, is partnering with August Home, a smart locks and smart home accessories provider, and same-day delivery company Deliv, to test the service.

Here’s how it works: Customers place their order online, and when the order is ready, a Deliv driver delivers it to the shopper’s home. If no one answers the doorbell, the driver enters a pre-authorized one-time passcode into a smart lock keypad installed beside the door. 

Customers receive a smartphone notification that the delivery is occurring, and they can monitor the delivery through home security cameras integrated with the August security app. Non-perishable items are left in the foyer, and fresh merchandise is placed into the shopper’s fridge. Once the Deliv associate leaves, the customer receives a notification confirming the delivery is complete and the door was automatically locked.

The concept is being tested among a handful of August Home customers in Silicon Valley. .... " 

Americans Loving Cooking?

As a long time amateur cook, found this very interesting.  Are the statistics shown right?   The aricle follows with an analysis of how this influences grocery choices.

The Grocery Industry Confronts a New Problem: Only 10% of Americans Love Cooking  by Eddie Yoon  In the HBR.

The supermarket and grocery business is likely to suffer strong headwinds in the future, due to long-term shifts in consumer behavior. Although many people don’t realize it yet, grocery shopping and cooking are in a long-term decline. They are shifting from a mass category, based on a daily activity, to a niche activity that a few people do only some of the time.

I’ve spent two decades consulting extensively for consumer packaged goods companies. Early in my career I gathered some data for a client on cooking. This research found that consumers fell into one of three groups: (1) people who love to cook, and cook often, (2) people who hate to cook, and avoid that activity by heating up convenience food or outsourcing their meals (by ordering out or dining in restaurants), and, finally, (3) people who like to cook sometimes, and do a mix of cooking and outsourcing, depending on the situation. At the time, the sizes of the three respective groups were about 15% who love to cook, 50% who hate to cook, and 35% who are so-so on the idea. ... " 

Personality Classification

Been examining various personality classifiers, starting with a project we did with the Watson Bluemix system.  Here is another, covered in some detail in  Datanami:   Deep Learning Reveals New Insights About People:

" ... Can a computer detect an author’s personality type, based only on a sample of his or her writing? Four researchers from Singapore and Mexico City sought to answer that question.

In their newly published study, the authors present a deep learning-based method that can figure out the psychological profiles of authors. They used a specially designed deep convolutional neural network. Their method analyzed various texts in order to identify the presence or absence of the Big Five personality traits. The full paper can be found in IEEE Intelligent Systems, a peer-reviewed academic journal. .... " 

Paper abstract and full text access.

See personality tag below for previous coverage of this topic.

B to B Sales and Consolidated Marketing Platforms

From CustomerThink

The Salesman’s Resurrection by AI Featured Column by CustomerThink Advisor!  by Thomas Wieberneit

The B2B salesperson is dead. .... Or so it is said.

There are scores of studies telling us that most of a customer’s buying process is already accomplished before a salesperson is even contacted and is merely allowed to make the transaction.
Forrester Research predicted in 2015 that 1 million US salespeople would lose their jobs to self-service and ecommerce by 2020. While they updated this report in March 2017 the statement largely stays the same.

The highly reputed Edelman Trust Barometer reveals in 2017 that trust in businesses declines and that “CEO credibility decreased the most, dropping to an all-time low”. Instead, people trust a person like themselves, or technical and academic experts most.

And the trust in experts is declining, too.

The Environment Has Changed

Today, enabled by technology, especially mobile devices, and social media, customers have a far higher reach than they had, say, in the 90’s. Today they can get a wealth of information online, via sources they trust. And they reach out for it, especially when it comes to products of low complexity. Especially for these products they can’t be bothered contacting a sales person at all but simply buy online.

Customers simply go through the process in their own time and at their own pace.

.... Towards a Consolidated Marketing Platform  driven by AI  ....  " 

Thursday, September 21, 2017

Streamlined Product Codes

Used to work with the GMA in our Innovation efforts.  Not in the business anymore,  but work ongoing is worth understanding as it interacts and communicates with consumers.

Watching a Good Idea on Streamlined Product Code Date Labels Become a Reality
By: Jim Flannery, Senior Executive Vice President, GMA and Mark Baum, Senior Vice President of Industry Relations, FMI

There is an undeniable satisfaction that comes with seeing a really good idea move from the realm of the imagined and into the concrete world of reality.  The grocery industry’s Product Code Date Labeling initiative – with its goal of helping consumers better understand and use produce date labels – is beginning to elicit that glow of gratification.

The problem? A myriad of different product date labels has left consumers confused on what to do with products in their homes. These package labels seek to provide important information about usage of products. However, product labels currently include a wide range of phrases such as Sell By, Best Before, Expires On, Use By, Best By, and Enjoy By. This varying phrases can confuse consumers and lead to disposal of food and products that are still fine to eat or use. .....  " 

GE on 3D Printing

Much more in the full article at the link.

A silent revolution: Avio Aero and Arcam’s relationship is built on additive manufacturing  by The Bike Shop

Less complexity, costs, waste and noise: additive manufacturing is the industry of the future and it’s already making history today.

Transformations that take several years and considerable time to bring about major changes are more commonly defined as progress, evolution or innovation. However, this does not deny their transformational nature: this comes to mind when considering what is happening today in the world’s largest industrial conglomerate, GE.

Just look at what is happening particularly in the technology known as additive manufacturing, also known as 3D printing. .... " 

Dynamical Machine Learning

Good piece, which combines process and sensors and IoT. Much more detail at the link. Reminds me comments made here before, about how we used classic process control in similar ways, and tried to link it to AI.  As long as it was a closed loop system, which few systems are, but if you can control the degree to which it is.  ..... There is the challenge and often the value in trying.

IoT Data Science & “DML” – match made in heaven?
By PG Madhavan  in DSC

DML stands for “Dynamical Machine Learning” (more in the book, “SYSTEMS Analytics for IoT Data Science”, 2017). This match is not surprising once you realize that DML & IoT are both based on the venerable Systems Theory. Let us dig deeper . . .

Consider IoT for industrial applications. A machine is instrumented with sensors, data are collected in real-time (or at intervals), communicated to the cloud where IoT Data Science techniques predict machine condition which results in an action, if necessary, such as repair action on the machine. This is a classic “closed-loop” system. The theory that abstracts and governs this closed-loop system is the subject matter of Systems Theory, an undergraduate engineering topic.

Systems Theory is broad and deep – in the past 70 or so years, a great body of work has been developed from deep theory to day-to-day applications such as GPS in your mobile phones, controlling massive chemical plants or Dreamliner airplanes. Systems Theory’s state-space model based methods allow you to describe, estimate/predict and control all parts of a closed-loop system. ... "

Solid State Cooling from SRI

Fascinating development reported in IEEE Spectrum.  Have long been a minor student of the physics of refrigeration,  will this development be practical for cooling or even refrigeration?   Also a mention of SRI, who we worked with over the years.  In fact SRI was long ago responsible for detergent innovations that we used and continue to leverage.    See SRI tag below for more mentions I have made, I will do a quick search of their work and report back.

" .... Can you imagine an electric cooler compact enough to fit in your pocket and flexible enough to wear? If not, think again because engineers at the University of California at Los Angeles and SRI International have one working: A 5-millimeter-thick device that is the world’s first solid-state cooler combining practicality, energy efficiency, and high performance. ...  "

Wednesday, September 20, 2017

Best Buy Banks on Smart Homes

A natural step forward, and a place to connect beyond entertainment.  Selection, design, installation, training, security ....  services come into play with devices and software.

Best Buy Banks on Smart Homes, Experiential Merchandising in Stores
By Adrianne Pasquarelli

In its quest to better connect with customers, Best Buy is focusing on connected living. At the Minneapolis-based retailer's Investor Day on Tuesday, its first in five years, executives laid out a new strategy of priorities and changes in its merchandising, training and marketing over the next three years.

Called Best Buy 2020, the plan is CEO Hubert Joly's second chapter in the growth plan he began in 2012, when he began transforming the 51-year-old retailer by zeroing in on the challenge of showrooming and opportunity in ecommerce.

Now, Joly is doubling down on smart homes, assisted living for seniors and in-home tech support. By the end of next month, the company plans to offer smart home shops and enhanced merchandising within all of its more than 1,000 stores. .... " 

Machines Learn to be Curious

Clever Machines Learn How to Be Curious 
Quanta Magazine  by John Pavlus

In Quanta.  Machines Learn to be Curious.
The University of California, Berkeley's Pulkit Agrawal is embedding curiosity, or intrinsic motivation, into artificial intelligence (AI) so machines can learn unfamiliar tasks more efficiently. "You can think of curiosity as a kind of reward, which the agent generates internally on its own, so that it can go explore more about its world," Agrawal says. His team at the Berkeley Artificial Intelligence Research Lab have developed a learning agent with an intrinsic curiosity module (ICM) so it can learn to play a video game. Their model is young children's innate curiosity in objects that surprise them, mimicked by the ICM's generation of an intrinsic reward signal defined by how mistaken the prediction model is. Agrawal says this reward for being surprised creates a feedback loop enabling the AI to correct its ignorance. The AI also translates visual input from raw pixels into an abstract version of reality, highlighting environmental features with the potential to affect the agent.   .... " 

AI Joins the Meeting

Attentive systems for meeting enhancements.

When AI Joins Your Meeting
By Rowan Trollope  

One of the best things about working at a company with Cisco’s resources is you get to blend fundamental technology research with macroeconomic trends. For example, the inevitable improvement of A.I. and collaboration technology, empowered by the global rise towards building distributed work teams, will lead to a revolution in how A.I. affects teamwork.  ... "

Coca Cola, AI and Big Data

The Amazing Ways Coca Cola Uses Artificial Intelligence And Big Data To Drive Success    By Bernard Marr 

" ... In fact, Coca Cola was one of the first globally-recognized brands outside of the IT market to speak about Big Data, when in 2012 their chief big data officer, Esat Sezer, said “Social media, mobile applications, cloud computing and e-commerce are combining to give companies like Coca-Cola an unprecedented toolset to change the way they approach IT. Behind all this, big data gives you the intelligence to cap it all off.”... 

More recently, Greg Chambers, global director of digital innovation, has said “AI is the foundation for everything we do. We create intelligent experiences. AI is the kernel that powers that experience.” ... ' 

Foot Traffic Counting

This company called Density was brought back to my attention for possible analysis.  we did this kind of analysis in the retail aisle.  (Their site at Density.io  no longer exists, leaving this here to state the problem)

" ... Density is, as the company’s website puts it so simply, a people counter. Like other startups that analyze foot traffic, Density is designed to let retailers, coffeeshops, co-working spaces and other places collect better data about how many customers pass through their doors.

In conversations at online forums, Density’s founders are candid about their belief that hardware sales are becoming a tough game. That’s why the company’s own device is both simple and free. It’s essentially just a pair of infrared sensors, with parallel beams crossing a doorway so it can count the people passing through and know whether each one is coming or going. Getting and installing the hardware costs nothing; Density makes its money through subscriptions to its data-analysis service. ... " 

Amazon Looking at Smart Glasses

Ultimately there will be non obtrusive, wearable communications devices in common use.

Report: Amazon is working on smart glasses powered by Alexa   in FastCompany

Amazon’s first wearable would tether wirelessly to a smartphone and look like a regular pair of glasses, avoiding the fashion pitfalls of Google Glass, reports the Financial Times. .... " 

Also in Technology Review.

Digital Factory Showing an Agile Way

Podcast on Digital Factory Model

Companies are beginning to use digital factories as incubators of more agile ways of working, often filtering the best attributes of the factory culture back to the larger organization.

A digital factory often calls for a whole new set of rules, including increased agility, new technology solutions, and cross-functional teams. Those differences have often spelled success for companies trying to develop and push out new digital capabilities quickly. In this episode of the McKinsey Podcast, senior partner Rohit Bhapkar and partner Joao Dias speak with McKinsey’s Barr Seitz about the challenges leaders might face in building digital factories and how to set themselves up for the best results. .... " 

Tuesday, September 19, 2017

Alexa Powered Lamp by GE


Had heard of this late last year, it is here:

GE's Sol lamp is the world's first Alexa-powered desk lamp, and it's brilliant  ... "

Amazon's Alexa voice assistant can no longer be contained to the company's Echo devices.

From smartwatches to Echo clones like the Eufy Genie, device makers are eager to integrate Alexa voice controls into everything imaginable. Late last year, GE announced the C by GE Sol, the world's first Alexa-powered smart LED lamp that requires no additional hardware or setup, and now it's finally available.

You plug it in, connect your Amazon account, and — boom — you're living in the future.  ... " 

Only $200 for preorder.

AI is Marketing Hype?

My team and I were involved in the last big hype cycle of AI.  We lost a large chunk of investment in the space.   So the question is a reasonable one.   What should we do?  How should we invest?

  There have been very big advances that we need to take note of, and use, but these have not launched us into an era of general artificial intelligence.  That remains in the future.   The applications of today are still narrow,  but clearly very valuable.    You cannot ignore them.  I am applying them now.  But it will need another jump forward to get real AI.  When?  I believe still decades away.

Is A.I. Just Marketing Hype?
Why are the big breakthroughs always five to 20 years in the future?   By Geoffrey James in Inc ....   "

Emergence of Practical Augmented Reality (AR)

I am a long time experimenter with the concept of AR in Business.  We experimented with very early (1980s) applications with specialized hardware.  For things like manufacturing line maintenance. Packaging analysis.   Applications were too cumbersome and narrow to be successful.  There was also a flurry of AR apps without extraordinary hardware assistance, also not very impressive for practical enterprise applications.

Now we are starting to see AR being embedded in real phones, with the potential for value beyond gaming. Below a list of ARKit enabled apps for the iPhone in IOS 11.  Beyond games.  Will be taking a look.

iOS 11 ARKit apps are appearing at the App Store
By Jonny Evans, Computerworld

Now that Apple has put reality distortion inside every iPhone, we’ll get to explore new realities with apps like these.   While Face ID may be capturing headlines, its Apple’s ARKit that will make a much bigger difference to iPhone user’s lives.

AR-ready apps set to launch

The company directed attention at gaming during its recent iPhone X launch, but entertainment apps are not all you should expect as the first ARKit-enabled apps appear.

Here are some of the first wave of AR-ready apps to be made available at the App Store around the launch of iOS 11. More are appearing rapidly at time of writing.  ... " 

Wal-Mart and in Aisle Checkout

Continue to follow the idea of in-aisle checkout, see my tag below.

Walmart deploying tech that lets in-store shoppers check out — on their own    By Deena M. Amano-MCCoy

Walmart is expanding the deployment of its Scan & Go mobile app, which is being tested in approximately 12 stores across Northwest Arkansas, Florida, Texas and Georgia. The chain is now rolling out the app for use in at least 10 additional locations in the Dallas-Fort Worth and Nashville markets. The stores will offer the solution by the end of the month, said Walmart spokesman Justin Rushing in an interview with Chain Store Age. .... " 

Nike Customizing Sneakers

Customization or perhaps late stage differentiation.  But note only for a limited time and place.  Plays to the current craze for fast delivery.  Adds a level of exclusiveness.  Discussion at the link too.

Nike customizes shoes for ‘sneakerheads’ in under an hour
 by Matthew Stern

Sneaker customization is a growing trend, one that appeals both to the fashion sense and the collector-mindedness of those shoe fanatics sometimes known as “sneakerheads.” Now Nike is offering a limited number of enthusiasts the chance to try out a new, fast form of customization technology.

Nike’s limited-time, invite-only event, called the Makers’ Experience, is taking place in the Nike By You Studio in New York, according to Engadget. Visitors choose from four possible packs of graphics and a few different color schemes to customize the upper portion of the shoe. They can also input custom text. The design is then projected onto a blank pair of sneakers the customer is wearing to demonstrate how the design will look. Once the final design is chosen, it takes an hour or less to create the final product. .... "


Constructing an Amazon Echo Skill via Java

How to construct an Amazon Echo Skill using Java    An introduction in Github.

Amazon, Robotics and Jobs

Amazon Just Showed Us What Really Happens When Robots Replace Jobs

They've already got 100,000 robots and are adding (a lot) more.
  
 By Lisa Calhoun   in Inc.  Video below. 

There are as many opinions about jobs and artificial intelligence as there are people who have them. A Vanguard global economist recently said AI will replace 1in 2 jobs. A new survey of AI researchers said most believe AI will outperform people at rote work like translating, surgery and transportation in 10 years. Fighting fire with fire, Google is offering a new AI job search. Marc Andreessen, founder at top venture capital firm Andreessen Horowitz, says humans have always been afraid whatever latest tech would take their jobs--and it's never happened. Meanwhile, some of the most automated factories in the country--like Tesla's gigafactory--can't hire fast enough because the demand for the product outpaces even well-automated labor.

Amazon is giving us the biggest real life case study about what happens when jobs--in this case, warehouse jobs--are replaced by robots. According to a recent New York Times article, Amazon has about 100,000 robots already. You can get a glimpse of what this looks like in this Youtube video of an Amazon fulfillment center .... " 

Monday, September 18, 2017

Open AI Framework by Microsoft and Facebook

From the Facebook research blog.  A further indication of the progress in standardizing how such systems should be built and maintained.   Would be useful to now compare this to work by Google in TensorFlow.   Is it comparable?

Facebook and Microsoft introduce new open ecosystem for interchangeable AI frameworks

By: Joaquin Quinonero Candela

Facebook and Microsoft are today introducing Open Neural Network Exchange (ONNX) format, a standard for representing deep learning models that enables models to be transferred between frameworks. ONNX is the first step toward an open ecosystem where AI developers can easily move between state-of-the-art tools and choose the combination that is best for them.

When developing learning models, engineers and researchers have many AI frameworks to choose from. At the outset of a project, developers have to choose features and commit to a framework. Many times, the features chosen when experimenting during research and development are different than the features desired for shipping to production. Many organizations are left without a good way to bridge the gap between these operating modes and have resorted to a range of creative workarounds to cope, such as requiring researchers work in the production system or translating models by hand.

We developed ONNX together with Microsoft to bridge this gap and to empower AI developers to choose the framework that fits the current stage of their project and easily switch between frameworks as the project evolves. Caffe2, PyTorch, and Cognitive Toolkit will all be releasing support for ONNX in September, which will allow models trained in one of these frameworks to be exported to another for inference. We invite the community to join the effort and support ONNX in their ecosystem. Enabling interoperability between different frameworks and streamlining the path from research to production will help increase the speed of innovation in the AI community. .... " 

Quartz Guide to AI

 Appears to be a guide useful for the non technical executive.   Good definition of terms at the end.  Intriguing view of the recent progress of AI and why companies care.  Reviewing.

Quartz Guide to AI    Don't Fear the AI ...

The Quartz guide to artificial intelligence: What is it, why is it important, and should we be afraid?
What is artificial intelligence? Why is it important? Why is everyone talking about it all of a sudden? If you skim online headlines, you’ll likely read about how AI is powering Amazon and Google’s virtual assistants, or how it’s taking all the jobs (debatable), but not a good explanation of what it is (or whether the robots are going to take over). We’re here to help with this living document, a plain-English guide to AI that will be updated and refined as the field evolves and important concepts emerge.  ... " 

Getting Design Tips from the Brain

Always interested in the choices made in bio mimicry of any kind.   The architecture will be key.

Computers Are Taking Design Cues From Human Brains 
The New York Times   By Cade Metz

Researchers are taking biological cues to remodel computing after the human brain to overcome the physical limits of semiconductors. "The existing [computing] approach is out of steam, and people are trying to re-architect the system," says former Stanford University president John Hennessy. Newer machines are no longer channeling all tasks through a single chip, but instead fragmenting them into smaller jobs distributed among farms of less power-consumptive, specialized chips. Former U.S. Defense Advanced Research Projects Agency program manager Gill Pratt expects an artificial intelligence "Cambrian explosion" to come from this trend, as the specialized low-power chips operate similarly to the brain. This architecture is yielding significant milestones, such as Microsoft's creation of a neural network with superior speech-recognition capability compared to people. Using graphics-processing units and specialized chips to respectively teach neural networks tasks and perform them could potentially help devices eventually accommodate more, and more complex, operations on their own.   ... ."   (May require Subscription) 

Interiew with Mobileye for Driverless

Getting the data right is the most important part.

Vision Sensors by Mobileye.  (Now part of Intel)    An interview.

" ... Though Google has become the US face of the driverless car movement, other companies have been developing similar technology for more than a decade. Mobileye is one of them, with a $10 billion valuation and a huge head start in a potentially enormous market. Professor David Yoffie discusses why a company many have never heard of will be a linchpin in the future of self-driving automobiles. ... " 

Google Expands Listening Research

A movement towards more voice driven assistance?  As a Google Home user I have not seen very much in changes there.   Still fairly narrowly applied.

How Google wants to build a world where every device is listening in TechRepublic   By Nick Heath

The firm is gathering speech samples from people across the globe, as part of a push to get simple voice recognition everywhere, paving the way for voice commands throughout our homes. ... " 

New Designs in Delivery Drones

How will these influence changes in the supply chain.
Designing Delivery Drones,   with video.

Odd Delivery Drone Design Carries Cargo While Keeping it Safe
By Dyllan Furnes in DigitalTrends

As delivery drones become more commonplace, this new design may help keep packages safe and secure on transit.

Researchers at EPFL have designed an origami-inspired delivery drone that could someday carry cargo, while keeping it safe in a cage along the way.

When unfurled, the drone can fly above buildings and traffic, while carrying a package up to 500 grams (a bit over one pound). When folded up, the drone is small enough to fit into a backpack.

Developed in EPFL’s Laboratory of Intelligent Systems, the delivery drone was designed to combine two functions. On one hand, the drone transports cargo; on the other hand, it serves as a protective packaging, acting like a resilient cage to keep cargo safe during its transit. .... " 

AI System Behavior

Teaching AI Systems to Behave Themselves
The New York Times  by Cade Metz

A small but growing community of artificial intelligence (AI) researchers is beginning to explore mathematical techniques that aim to prevent machines from demonstrating unexpected, unwanted, and harmful behavior. Safety concerns are growing as AI moves into online services, security devices, and robotics. Researchers at OpenAI are developing algorithms that can learn tasks through hours of trial and error while receiving regular guidance from human teachers. The researchers believe algorithms that combine human and machine instruction can help keep automated systems safe. OpenAI is working to build reinforcement-learning algorithms that accept human guidance to ensure that systems do not stray from assigned tasks. Some researchers in the field are working to ensure that systems do not make mistakes on their own, while others aim to prevent hackers and other bad actors from exploiting AI flaws.  ,,, " 

Podcast on Lego

Have followed Lego for years.  Both as a creative toy, and as a means to model more generally.

How Lego Can Rebuild Its Business

Wharton's David Robertson discusses what went wrong at Lego.

Danish toymaker Lego recently surprised its industry watchers when it announced plans to trim its 18,200-strong workforce by about 8% to shed about 1,400 jobs after seeing revenue drop for the first time in 13 years. Lego said revenue fell by 5% to Danish krone 14.9 billion (US$2.4 billion) in the first half of 2017 compared to the same period last year as sales weakened in mature markets such as the U.S. and parts of Europe. The drop is a shock for the toymaker, whose annual revenue nearly quintupled between 2007 and 2016 to DKK 37.9 billion (US$6 billion). But the warning signs were clear when revenue growth slowed from 25% in 2015 to 6% in 2016.

Lego said it would take steps to simplify operations that have become increasingly complex as a result of double-digit growth. “We have added complexity into the organization which now makes it harder for us to grow further,” said Lego chairman Jorgen Vig Knudstorp. “We have now pressed the reset-button for the entire group.” Lego will build a “smaller and less complex” organization to simplify its business model and reach more children. In some markets, Lego will clean up its inventories across its entire value chain. .... " 

Sunday, September 17, 2017

Towards a loss of Anonymity

Face recognition app taking Russia by storm may bring end to public anonymity ... FindFace compares photos to profile pictures on social network Vkontakte and works out identities with 70% reliability ...  " 

On Managing Cyber Risk

Of interest:

Security Researcher at UpGuard, Roane Holman 

The Executive's Guide to Managing Cyber Risk  (Requires registration) 
Regardless of what stage of digital transformation an enterprise finds itself at, the same risks and challenges await at every turn– namely, in the areas of security, compliance, and data management. Learn actionable strategy with this new report...  

Just How Smart are Smart Machines?

Very good piece from some of our communications with MT Sloan.  Read the whole thing.  May require some registration.   Again, very non technical, management  approach to the question.

Just How Smart Are Smart Machines? 
Thomas H. Davenport and Julia Kirby

The number of sophisticated cognitive technologies that might be capable of cutting into the need for human labor is expanding rapidly. But linking these offerings to an organization’s business needs requires a deep understanding of their capabilities.

Smart Machines
If popular culture is an accurate gauge of what’s on the public’s mind, it seems everyone has suddenly awakened to the threat of smart machines. Several recent films have featured robots with scary abilities to outthink and manipulate humans. In the economics literature, too, there has been a surge of concern about the potential for soaring unemployment as software becomes increasingly capable of decision making. Yet managers we talk to don’t expect to see machines displacing knowledge workers anytime soon — they expect computing technology to augment rather than replace the work of humans. In the face of a sprawling and fast-evolving set of opportunities, their challenge is figuring out what forms the augmentation should take. Given the kinds of work managers oversee, what cognitive technologies should they be applying now, monitoring closely, or helping to build? ... " 

Output Visualization for Analytics

Good practical piece out of the latest  Informs Analytics Magazine.   They still prefer to call it analytics rather than data science.    And its always been 'Machine Learning', though not called that. You have always learned from your data via machines, software and data

Output visualization of machine learning analysis
By Navneet Kesher

Data science is more than just building machine learning models; it’s also about explaining the models and using them to drive data-driven decisions. In the journey from analysis to data-driven outcomes, data visualization plays a very important role of presenting data in a powerful and credible way.  .... " 

This Blog on Twitter

From nearly the beginning, this blog has been connected to Twitter.   You can follow it there from my handle @FranzD.  I  selectively mention many posts here.   Some people find that simpler to follow.  See: https://twitter.com/franzd

Wineries of the Future

Here a big proponent of digital makeovers, though I do wonder when basic physical process should be replaced by automation.   Impressive attempt here I had not heard of.  A control room that looks like a nuclear plant.    Good description here, with lots of images.

A look at the Winery of the Future     by Sam Slaughter in Digital Trends.

When winemakers a century from now look back on the technology that changed the game, one winery is going to stand out: Palmaz Vineyards in the Napa Valley. Though ordinary in appearance at first glance, an amazing array of custom-built technologies lurking in a subterranean lair bolster the winemaker’s craft with exacting science.

Located on 610-acre parcel located northeast of downtown Napa, the vineyeard is run by the Palmaz family, who originally came from Argentina. They purchased the property in the mid-90s and had to start from scratch, as the previous owner had let the ground go fallow.

When they set out to make the winery, the Palmaz family wanted to take the land into account. They drilled 4,900 core samples around the property to understand the various soil compositions, then tailored the entire layout of the vineyard around the findings – a departure from the simple rectangular parcels that dominate the surrounding area.

Another goal was to have access to as many available options at the time of blending as possible.

“What if we could build a facility to leverage all these unique regions and never allow them to be combined until the moment of blending? Then, to take that one step further, never make consolidations of any kind,” Palmaz says. Take these two elements, throw in three years of planning, seven of construction, and two of finishing touches, and a winery was born.   ... " 

Saturday, September 16, 2017

The Pickle Index

Always liked the idea of a non-sequential novel.  It has been tried many times, never successfully for my taste.  Is it here now?

No relation.  InWired: 

The Pickle Index is a delightfully Weird, App-Driven Novel like no other. .... 

Through The Pickle Index—an app, a hardcover, and a paperback book—users enter into Destina, a dystopian world governed by pickle consumption. .... 

The fact is, The Pickle Index is not a traditional novel, nor is it a conventional app. When Eli Horowitz and Russell Quinn set out to create the multimedia storytelling experience, they made a conscious decision to eschew hallmarks of design like accessibility and ease of use. Instead, they provide multiple entry-points into an intricate and immersive world. In doing so, they’ve reimagined what a digital literary experience can be. .... " 

Algorithmically Determining Justice

In the Alumni Mag, a considerable, but non technical piece that addresses the use of analytical methods to automate crime detection and apply penalties.   Moving closer to the world of the minority report?

The Pennsylvania Gazette
Black Box Justice   by Trey Popp     28 Aug 2017

Richard Berk designs computer algorithms that predict crime. As courts and cops increasingly use his and similar tools to shape everything from parole decisions to street policing, Berk has a warning: accuracy comes at the cost of fairness, and citizens must decide where justice lies.  .... " 

Hive Robotics

Swarming robotics to perform tasks collaboratively is something we examined in warehouse applications.  A different, but structurally similar problem came up recently and I did some searching for the state of the art. This example came up.  A way to mock up the idea at least.   See the video linked to:

In these bot hookups, the machines meld their minds    In Science News: 

New composite robots take coordination to a new level   By Maria Temming   Meet the first fleet of hive-minded robots.

These robots can latch onto one another and link up their individual “nervous systems” — the network of connections between their computer processors, cameras, wheels and other gadgetry — to create a single-minded machine. The composite robots, reported online September 12 in Nature Communications, pave the way for a new generation of machines that can change their shape on the fly.

In previous robots made up of many smaller machines, each robotic building block maintained control over its own nervous system. It was “a little bit like if we had a bunch of people joining together to do something,” explains study coauthor Marco Dorigo, an engineer at Universit√© Libre de Bruxelles in Belgium. “They were trying to coordinate by exchanging messages with neighbors without anyone being in charge.” These multi-bodied robots could link up to create different shapes, unlike solo devices. But the lack of centralized control meant the robot collectives were clumsier and had slower reaction times than stand-alone bots.  .... " 

Considerations with Combining AI and Data

An interesting and lengthy piece in InfoQ on key issues with the construction and delivery of AI solutions. Use of data ontologies are mentioned,  a good idea.   Their key takeaways are instructive:

- Machine learning in turn is frequently fueled by big data but can also be fueled by traditional data sources.

- No matter what the scope is, we have to select data that is appropriate to the domain of the problem space

- Information from highly diverse sources needs to be parsed, curated, packaged, contextualized, and componentized for consumption by users or ingested by systems.

- While machine algorithms play an important role in both the preparation of data and interpretation of user intent, these types of applications require a significant amount of knowledge engineering to be successful.

- Thinking about data as a service and the platform as an orchestration layer between business problems and technology solutions can help organizations achieve dramatic improvement in data scientists productivity. .... " 

Much more at the link.

AI Protecting Us

This reminds me of the many 'maintenance'  type systems we worked with.  Warnings can be very useful.  Then add a recommendation tailored to the situation.  And a dialog for the details of  implementation.

Treating Cancer, Stopping Violence . . . How AI Protects Us
By BBC News 

For some, the spread of artificial intelligence and robotics poses a threat to our privacy, our jobs – even our safety, as more and more tasks are handed over to silicon-based brains. .... " 

Groups Lie More than Individuals

In Analytics Magazine:

" ... When organizations are exposed for large-scale deceptive or corrupt behavior, often it is not the actions of one or two employees, but a coordinated effort of many individuals, to include upper level management. Prominent examples include the bankruptcies of WorldCom and Enron, and even more recently, the alleged issuance of faulty emissions certificates by German car manufacturer Volkswagen. The study, “I lie? We lie! Why? Experimental Evidence on a Dishonesty Shift in Groups,” explores what motivates a group of people, especially those who previously behaved honestly, to work together to deceive.

The study authors, Martin G. Kocher, Simeon Schudy and Lisa Spantig, all of the Ludwig-Maximilians-University of Munich, studied 273 participants in both individual and group situations. Participants, who were paid for their role in the study, were shown video of dice rolls and asked to report the number shown on the die. The higher the reported die roll, the larger the monetary compensation. Participants were evaluated on an individual basis, and in two group settings: one in which all members of the group must report the same die roll to receive a payoff, and another in which members do not have to report the same die roll to receive a payoff. In the group settings, members are able to communicate with each other via a chat feature.

“We observed that groups lie significantly more than individuals when group members face mutual financial gain and have to coordinate an action in order to realize that financial gain,” Kocher says. ... " 

Friday, September 15, 2017

First lets Replace all the Lawyers

From the O-Reilly Newsletter:  (At the link there are lots of links for more detail)

DoNotPay steps into the Equifax breach

The Equifax data breach could expose about half of the US population to risk of identity theft, and—as has been widely reported—the credit-reporting company has bungled its response in many ways. Joshua Browder, a Stanford student who initially created the DoNotPay robolawyer to help people fight parking tickets, has already expanded its powers to help homeless people find housing and refugees apply for asylum. He has now added a new wrong for the bot to right. DoNotPay is taking on Equifax, enabling people to sue the company in small claims court in all 50 states, for up to $25,000, at no charge for the bot's services. Browder explains how it works in this CNBC interview.

+ Watch: Will bots replace lawyers? Joshua Browder's talk at the 2016 Next:Economy Summit. 

+ Wall Street traders: It's your turn to start worrying about robots.  .... " 

Small Companies Competing with Watson

I am part of this ecosystem myself. Certainly the small company cannot  afford a complete installation from IBM.  But IBM has made open capabilities available via Watson Bluemix, so assembly work can be done by cheaper contractors,  or the work can be done internally.  IBM has also published a number of end to end tutorials for cognitive AI use.   And in a number of cases provided fully autonomous, focused domain systems like Watson Analytics.

Watch Out Watson: Small Companies Threaten IBM’s Artificial Intelligence
Innovative, smaller companies show that there can be a more cost-effective way to approach AI than complex systems like IBM's Watson. .... " 
By Teresa Riva

Marketing Skills for the Amazon Show

Key will be how new skills integrate with the capabilities of the device, and innovations of marketing, looking forward to see the possibilities.    Lots of video content out there,  but how will this be different from

Marketers find a new video platform: Amazon Echo Show
By Ilyse Liffreing in DigiDay

The Digiday Content Marketing Summit is coming up next month in Vail, Colorado, and we’re opening up a limited number of complimentary passes for content marketers from brands. Interested? Apply here.

When Amazon’s Echo Show officially launched on June 28, publishers rapidly announced video-optimized skills for the device, while Amazon only partnered with a handful of brands to do the same. Reviewers swiftly critiqued the device for not having enough visual skills.

But it seems the fever is building for the $230 audio and video device. More brands are now optimizing their Amazon skills for the new Echo Show.  .... " 

Oracle's Adaptive Intelligence for CX

Oracle is calling it adaptive intelligence, and is pushing it for better customer experience.

Oracle OpenWorld: Got Customers? Better Get AI For The Best Customer Experience    Linda Currey Post , Oracle in Forbes.

Better customer experiences will be a central topic at Oracle OpenWorld 2017, which takes place October 1 to 5 in San Francisco. Oracle experts, customers, and industry thought leaders will share how artificial intelligence (AI) technology is enhancing customer acquisition, retention, and loyalty by offering consumers the best possible interactions across all channels.

Oracle’s approach to AI, known as Adaptive Intelligence, combines several technologies, including data science, artificial intelligence, and machine learning. AI-powered systems can recognize images, understand language, learn from data, and emulate human intelligence.  Oracle’s recently introduced suite of Adaptive Intelligent applications is focused first on applying customer insights derived from AI to the transformation of customer experience, and will soon span the enterprise to include applications for finance, human resources, and supply chain management.   .... " 

Job Losses and Automation

Basically, we don't know the numbers, the complexity of the system involved is large.

Statistics on job losses due to automation are sensationalist: Anthony Schofield, Deloitte   By Prachi Verma

Thursday, September 14, 2017

A TensorFlow Tutorial


Have now seen several tutorials, this one starts simply   I will be following the other parts and give some commentary as I can.   For a definition of TensorFlow, see how Google describes it at their site:

 .... TensorFlow™ is an open source software library for numerical computation using data flow graphs. Nodes in the graph represent mathematical operations, while the graph edges represent the multidimensional data arrays (tensors) communicated between them. The flexible architecture allows you to deploy computation to one or more CPUs or GPUs in a desktop, server, or mobile device with a single API. TensorFlow was originally developed by researchers and engineers working on the Google Brain Team within Google's Machine Intelligence research organization for the purposes of conducting machine learning and deep neural networks research, but the system is general enough to be applicable in a wide variety of other domains as well. ... " 

Tensorflow Tutorial : Part 1 – Introduction

By Vivek Kalyanarangan on September 12, 2017  

In this multi-part series, we will explore how to get started with tensorflow. This tensorflow tutorial will lay a solid foundation to this popular tool that everyone seems to be talking about. The first part will focus on introducing tensorflow, go through some applications and touch upon the architecture.
This post is the first part of the multi-part series on a complete tensorflow tutorial –

Tensorflow Tutorial – Part 1: Introduction
Tensorflow Tutorial – Part 2: Getting Started
Tensorflow Tutorial – Part 3: Building the first model ..... " 

How China is doing Digital

Quite a good piece, with statistics of interest. Updated details here.

China’s digital economy: A leading global force

By Jonathan Woetzel, Jeongmin Seong, Kevin Wei Wang, James Manyika, Michael Chui, and Wendy Wong

China is already more digitized than many observers appreciate and has the potential to set the world’s digital frontier in coming decades.

China has one of the most active digital-investment and start-up ecosystems in the world, according to a new discussion paper from the McKinsey Global Institute (MGI), China’s digital economy: A leading global force. China is in the top three in the world for venture-capital investment in key types of digital technology, including virtual reality, autonomous vehicles, 3-D printing, robotics, drones, and artificial intelligence (AI). China is the world’s largest e-commerce market, accounting for more than 40 percent of the value of worldwide e-commerce transactions, up from less than 1 percent about a decade ago. China has also become a major global force in mobile payments with 11 times the transaction value of the United States. One in three of the world’s 262 unicorns (start-ups valued at over $1 billion) is Chinese, commanding 43 percent of the global value of these companies (Exhibit).  .... " 

Kroger Tests Kitchen 1883 Restaurant Concept

Had seen this done on the west coast, apparently with some success. ( At first I thought this would be in store or by store.  But that is unclear from the press release ) .  Comfort food makes sense, with regional and ethnic contributions. How can this be best linked to the store?

The Kroger Co. Introduces Restaurant Concept Kitchen 1883

CINCINNATI, Sept. 14, 2017 /PRNewswire/ -- The Kroger Co.'s (NYSE: KR) Culinary Development team today announced a new restaurant concept, Kitchen 1883, which offers a fresh take on new American comfort food. .... " 

http://www.kitchen1883.com/
First store in Union KY in Fall of 2017.

Neato Maps Rooms

More cleaning, mapping.  Other applications in the smart home?

Neato D7
Neato Adds Persistent, Actionable Maps to New D7 Robot Vacuum
By Evan Ackerman in IEEE Spectrum

The Neato D7 remembers the layout of your home, allowing it to recognize virtual no-go lines and respond to them as it cleans, doing away with physical barriers.

Vacuums are the most successful consumer home robots ever, and they’re likely to remain so for quite a while, simply because they do quite a good job at keeping your floors clean in a reliable and affordable way. For robot vacuum manufacturers, this is a bit of a problem: As we very often point out in our reviews, lower-end models clean just about as effectively as more expensive ones, often making it difficult to justify an upgrade.

Today, Neato Robotics is introducing a new flagship robot vacuum that we think offers one of the most significant advances we’ve seen in years: persistent, actionable maps. Like its predecessors, the D7 uses a lidar sensor to create a map of your house as it goes, but now, the robot will remember that map and allow you to interact with it. Neato is starting off simple with what you’ll be able to do (like defining no-go zones), but it’s an incredibly powerful feature that’s necessary for the future of all home robots. .... " 

Wednesday, September 13, 2017

Age of Digital Assistants

Not a bad view, but the longer term potential is greater than implied.

The Age of Digital Assistants – What and when to expect?       By Shishir Kathurwar  in CustomerThink

“ The better we get, the less time you spend talking to customer service. It’s a gain for companies, but it’s also a gain for personal life.”– Alex Lebrun, Facebook M’s teacher  ... " 

More on my ongoing look at and development of Virtual Assistants.

Transitioning to Industrial AI

Some quite useful thoughts here, about work flow and production versus testing, retesting and retraining systems. Ultimately we learn this, but sometimes painfully.  Not different from implementing any kind of analytics or solution improvement based system.  Its fairly rare that it is rigorously done.

Transitioning from Academic Machine Learning to AI in Industry
Jeremy Karnowski and Emmanuel Ameisen, Insight AI  .... " 

Waymo Training Cars

Inside Waymo's Secret World for Training Self-Driving Cars

An exclusive look at how Alphabet understands its most ambitious artificial intelligence project

Alexis C. Madrigal ....

Watson Analyzing Social Tone

Instructive Watson Example, via O'Reilly

How can I use IBM Watson to analyze the tone of social media posts?
Learn how to use IBM Watson's APIs and natural language understanding to analyze the tone of social media posts like tweets.  Video.

By Matt Coatney  ... " 

Airbags for Robots (To protect the Humans they Work With)

From Idea Connection.  Includes a video.  Via Ideaconnection. 

Airbags for Robots Protect Co-Workers 

Inflatable airbags for robots could help to protect their fragile, human co-workers.

As robots begin to work more closely with humans, there is also an increased risk of accidental injury. To help reduce that concern, researchers from Germany’s Institute of Robotics and Mechatronics developed a Robotic Airbag that would be inflated automatically as the “cobot” (collaborative robot) moves about at high velocities. This cushion would eliminate the need for extra safety sensors, while also allowing the co-bot to work at maximum efficiency. .... " 

Tuesday, September 12, 2017

Apple on Virtual and Augmented Reality

Saw these presentations today.  Relatively little was presented on AR and VR, except in the game space.  But apparently it will be presented live in their visitor center.   The iPhoneX seems to be built for it.  But the content available outside gaming is still thin.

Apple’s New Visitor Centre To Feature AR Experience

The centre will open later this year.   By Peter Graham   in VRFocus 

Today, Apple held a special press conference event to make several big product announcements. Being held for the first time in the new Steve Jobs Theater, located at Apple’s new headquarters in California, the event went straight into an early mention of augmented reality (AR) that would greet visitors. .... "

Reshaping Business with AI

Good executive view.

Reshaping Business with Artificial Intelligence

Closing the Gap Between Ambition and Action

by: Sam Ransbotham, David Kiron, Philipp Gerbert and Martin Reeves

Disruption from artificial intelligence (AI) is here, but many company leaders aren’t sure what to expect from AI or how it fits into their business model. Yet with change coming at breakneck speed, the time to identify your company’s AI strategy is now. MIT Sloan Management Review has partnered with The Boston Consulting Group to provide baseline information on the strategies used by companies leading in AI, the prospects for its growth, and the steps executives need to take to develop a strategy for their business.  ...  "

Request full article at the link, requires registration.

Privacy of Home Robotic Data

Thoughful piece that follows up on the use of home robots building models of homes, and potential  sharing of that data.     More broadly we will see this kind of data also being gathered by AR systems as well, for example Google's Project Tango.   In that case not robotic, but human 'scanning'.

Interview: iRobot CEO Colin Angle on Data Privacy and Robots in the Home    By Evan Ackerman   in IEEE Spectrum

About a month ago, iRobot CEO Colin Angle mentioned something about sharing Roomba mapping data in an interview with Reuters. It got turned into a data privacy kerfuffle in a way that iRobot did not intend and (probably) did not deserve, as evidenced by their immediate clarification that iRobot will not sell your data or share it without your consent .... 

Data privacy is important, of course, especially for devices that live in your home with you. But as robots get more capable, the amount of data that they collect will increase, and sharing that data in a useful, thoughtful, and considerate way could make smart homes way smarter. To understand how iRobot is going to make this happen, we spoke with Angle about keeping your data safe, integrating robots with the future smart home, and robots that can get you a beer.  ... " 

Clearpath Moderizes TurtleBot

Clearpath Modernizes TurtleBot With Intel Euclid and iRobot Create  By Evan Ackerman in IEEE Spectrum 

Clearpath Robotics, in partnership with Intel and iRobot, recently announced the newest member of the TurtleBot family: the TurtleBot Euclid. The new ROS-powered robot features an iRobot Create 2 mobile base along with a new Intel Euclid sensing and computing module. It's designed to be more powerful and less expensive than the original Turtlebot 2, and promises to be easier to use than ever.  ... " 

For prototype, experimental and teaching purposes.  Connection with iRobot and Intel are interesting.  Note that Intel is doing sensing capabilities,  continuing previous work.

Mapping Sound Environments

We experimented with managing sound environments in retail.  Perhaps a more rigorous way to look at the problem?   See my tags on this.

A New Mobile Application Helps Scientists Map the Sound Environment

via CRNS: 

".. Researchers at the National Center for Scientific Research (CRNS) in France and IFFSTTAR (the French institute of science and technology for transport, spatial planning, development, and networks) have developed NoiseCapture, an application created within the framework of the ENERGIC-OD European project that unites specialists in environmental acoustics and geographic information. NoiseCapture is accompanied by a comprehensive information system that can store, query, and share collected data. In addition, the system functions in real time, and will enable the creation of a permanent global database. The research teams want to open up research on the sound environment to the broadest possible audience. They have developed NoiseCapture as an "open science" project, made resources available under an open source license, provided the data they collected under an open data license, and published the resulting research in open access journals. The researchers want to use the NoiseCapture project to involve citizens in the control of their territory, especially in matters of noise pollution.  .... " 

Monday, September 11, 2017

Alexa as Companion for Hospital Stay

An example of AI and chatbot digital assistants being tested for hospital patient interaction.  See the tag below: 'Hospital Virtual Assistants'  for more diverse examples and coverage.   I will continue to use this tag in coming days.   Please pass along other examples you may have.

Alexa: 24/7 Companion for a Long Hospital Stay   by Richard Satterwhite

Patient Engagement Specialist, Office of the Patient/Family Experience ...  "  

Robotics in Hospital Operations

Continuing my look at Hospital Operations and AI/Robotics/Assistants,  especially with regard to patient interaction and comfort.

Adoption of Robotics Into a Hospital's Daily Operations Requires Broad Cooperation  VTT Technical Research Center

Researchers at the VTT Technical Research Center of Finland have analyzed deploying a logistics robot system at the Seinajoki Central Hospital in South Ostrobothnia, Finland, with the goal of cutting transportation costs, improving supplies availability, and mitigating congestion in hospital hallways via 24/7 deliveries. The researchers found the system's success relied on joint planning and dialogue between occupational groups and stakeholders. Of particular importance was accounting for the interdependencies between various players, along with their roles in the hospital's core task. The researchers note the smooth process of robotic adoption overall demands careful planning, piloting, and implementation. "As the system is expanded with new robots and types of deliveries, even more guidance, communication, and dialogue is needed," says VTT's Inka Lappalainen. "Joint planning that brings various players to the same table ensures that the system's implementation goes as smoothly as possible, making it easier to achieve the desired overall benefits." ...  ' 

Machine Learning Resources

Good list, mostly courses, on these topics.

Writing about AI/ML, growing @YupTechnologies. Previously @Upstart, @Yale, @TrueVenturesTEC.    By Vishal Maini 

The Best Machine Learning Resources
A compendium of resources for crafting a curriculum on artificial intelligence, machine learning, and deep learning. .... " 

Choosing Better Passwords

Just started to rethink the effective use of passwords here ....

Choose Better Passwords With the Help of Science 

in The Conversation
Lorrie Cranor; Blase Ur; Lujo Bauer, et al

A team of researchers has been conducting experiments into improving password security, using scientific methods so that such passwords are difficult for computers to guess. The researchers found many users often subscribe to misconceptions, such as thinking that adding a number or punctuation mark to the end of their password will strengthen it. They also say their findings indicate it is more important for passwords to be longer instead of more complicated, while long passwords still tend to be predictable. The researchers note providing feedback while people are creating new passwords can help significantly, and their password meter uses an artificial neural network to rate password strength based on analysis of other passwords. Among the team's recommendations are making passwords at least 12 characters long, and mixing in at least two or three different types of characters in unpredictable places. They also advise against reusing existing passwords for multiple, sensitive accounts. .... " 

Sunday, September 10, 2017

Webinar on AI Fueling the Retail Supply Chain

A.I.’s role in fueling the retail supply chain

Retail operators are at a Darwinian crossroads, facing a future in which evolution equates to survival.  Many retailers are looking intently at advanced cognitive science as a way to heighten their understanding of customer behavior and, in turn, optimize their business practices.

In this quest, A.I. holds great promise for a wide range of retail applications — from robotic production to predictive analytics and improving human-to-machine communications. As theory advances into retail practice, RetailWire, in coordination with IBM, sought an understanding of what aspects of the value chain retailers and their partners think are ripe for A.I. improvements.

See this on-demand, interactive event as we present the results of our survey of nearly 200 retail industry execs. We also interview a live panel of experts to get their interpretation of the study data and their advice for retail operators.

Feature presenter:
Jennifer Van Cise
Vice President, Industry Solutions & Business Development, US Distribution Market, IBM

BrainTrust Panelist:

James Tenser
Principal, VSN Strategies

Mark Heckman

Principal, Mark Heckman Consulting

Moderated by:

Al McClain
CEO, RetailWire

IBM and MIT Say Materials and Quantum will Supercharge AI

IBM and MIT Bet That Materials and Quantum Advances Will Supercharge AI 
Technology Review

By Will Knight; Elizabeth Woyke

The Massachusetts Institute of Technology (MIT) and IBM have announced a project to probe new deep-learning approaches and devices, materials, and physical phenomena to advance artificial intelligence (AI), including quantum computing. The partners will establish a new research facility based at MIT for hardware innovations, as well as new types of machine-learning algorithms. The latter area of concentration will explore algorithms that could enable computers to gain knowledge from unlabeled data, as well as those that could make possible the transfer of learning from one domain to another. "We're not going to be designing algorithms that are completely independent of the architectures we're going to be using," says Anantha Chandrakasan, dean of MIT's school of engineering. "We'll see system-level thinking." Chandrakasan also says the facility will investigate the practical application of AI in specific domains, including computer security and healthcare. ... " 

Chatbots, Theory and Practice

Very nicely done piece on Chatbots, and especially conversations that they claim to produce.  Non technical overview.

By Jonathan Mugan in Medium:
AI and machine learning. Co-founder and CEO at DeepGrammar, a startup specializing in NLP and deep learning. Author of The Curiosity Cycle.

Chatbots: Theory and Practice
There’s a lot of fluff surrounding chatbots, so I wrote this post to lay out the basics. I first review the theory of conversation to give us a sense of what we are aiming for. I then discuss three classes of chatbots. The simplest class is purposeless mimicry agents, which only provide the illusion of conversation. Members of this class include ELIZA and chatbots based on deep learning sequence-to-sequence models. The second and next most sophisticated class comprises intention-based agents such as Amazon’s Alexa and Apple’s Siri. These agents have a simple understanding and can do real stuff, but they generally can’t have multi-turn conversations. The third and most sophisticated class is conversational agents that can keep track of what has been said in the conversation and can switch topics when the human user desires.  .... " 

 I see that much of this is also covered in a Video.

See also DeepGrammar.com

Plantoids: Plant Robotics

I am a botanist from way back, so from time to time you will see tech oriented agri and horti cultural notes.   Here from Kickstarter, a still unavailable robotic plant pot that seeks out the best place to live in an environment.  Not very practical, but you might think of ways to have larger scale applications of the idea.  See previous examples of the Tertill plant tending robotic,   more practical I think, but scalability a big problem.

" .. Concept:

Our Plantoids are practical. We actually build plant brained robots (no more abstract models), and approach the educational topics from a mindset of building a model with real hardware and code that results in physical experiments. Your plant-robot cyborg dream can become reality. 

Sensory Input:

The key is using sensors and data to expand and support the survival of the plant specimen on the robot. When you can achieve that, the robot can perform interesting experiments and missions to expand life to new environments--previously devoid of life! 

Using an Arduino-based sensory suite to provide data on the quality of life of its plant center, a Plantoid can autonomously explore its environment in search of the best location for its own prosperity. Modeling the extension of life can lead to literacy that may inspire new solutions to big problems that we face today: send life to other planets, change food production, or model survival and sustainability in harsh environments. 

Currently, the sensors included in the Plantoids Base Kit:

Soil moisture
Air temperature and humidity
Ambient light
Air quality sensors.  .... " 


Brands engineering Social Media Content

Brands Engineering Social Media Content

Wharton's Kartik Hosanagar discusses his research on brand content and consumer engagement on social media.

In the world of social media advertising, the biggest win for firms is when consumers are delighted by the content they see, want to engage with it and eventually buy something. Kartik Hosanagar, Wharton professor of operations, information and decisions, has co-authored research that takes a closer look at brand posts on Facebook to determine the type and mix of content advertisers should aim for to get results. The paper, “Advertising Content and Consumer Engagement on Social Media: Evidence from Facebook,” which was co-authored with Dokyun Lee of Carnegie Mellon University and Stanford University’s Harikesh Nair, is forthcoming in the journal Management Science. Hosanagar recently joined Knowledge@Wharton to discuss his findings. ... 

An edited transcript of the conversation follows .... " 

Saturday, September 09, 2017

Library to Preserve Sounds

Usually don't think of sounds in general,  as opposed to music, but interesting to explore. The link has the ability to see a number of examples.

British Library exhibit to highlight the sounds it’s fighting to save
"Listen: 140 Years of Recorded Sound" runs from October 6th to March 11th next year.

By Jamie Rigg, @jmerigg

Quantum Bayesian Networks

Looks to be of interest, follow along:

Quantum Bayesian Networks   August 15, 2017

Resistance is Futile: Jupyter Borg Threatens to Assimilate Quantum Computing Universe
Filed under: Uncategorized — rrtucci @ 5:00 pm 

A week ago, IBM announced at its Quantum Experience usegroup that it had uploaded to github a large collection of jupyter notebooks exhibiting the use of their gate model quantum computer (previously 5 qubits, currently 16 qubits). I consider this an excellent addition to the quantum open source and free jupyter notebook universe and ecosystem. I’ve advocated for quantum open source and jupyter notebooks many times before in this blog, so it’s a pleasure for me to echo their announcement.  ...   " 

The Jupyter Notebook: 
The Jupyter Notebook is an open-source web application that allows you to create and share documents that contain live code, equations, visualizations and explanatory text. Uses include: data cleaning and transformation, numerical simulation, statistical modeling, machine learning and much more .... " 

Jerry Pournelle Passes

Jerry Pournelle passes, an early adopter of computing technology, " ... the first author to write a novel on a computer ... " ,  science fiction writer and journalist.   I was a follower from the beginning of personal computing.   More in the Verge.  And in the Wikipedia.

Ultimate Customer Experiences

How AI, Machine Learning & Data Visualization are Defining the Ultimate Customer Experience      by Philip Piletic 


The words artificial intelligence (AI), machine learning (ML) and data visualization are everywhere right now. Both AI and ML have gained an immense role in defining the business world and have especially influenced the way we define the customer experience.

But when it comes to understanding AI and ML, some people tend to mix them up or they believe that one can substitute the other, which is not exactly the case.

The difference between AI and ML
This term is used in order to define technologies that have a human- like intelligence, which are able to do cognitive tasks that were usually reserved to human minds. The term machine learning is meant to describe the process machines go through in order to get data and basically learn for themselves. A better understanding of advanced applied analytics, which is what both AI and ML are based on is definitely necessary in order to bring improvements to all business areas.

To paint a picture of how far AI has come, we bring up the victory of Google’s AlphaGo against Go grandmaster Lee Sedol, in March 2016. Considering that Go is a game that requires strategic thinking, creativity and, not to mention, intuition, it’s easy to understand that when AlphaGo won the game, it also reached an important milestone for such a program.  ... "

Friday, September 08, 2017

Sequence Prediction Problems

A handy tutorial on the use of machine learning to do sequence prediction, by Jason Brownlee.  A kind of elementary forecasting.   Nicely done.    Read his piece and sign up for more of his services:

'  ... Sequence prediction is different from other types of supervised learning problems.  The sequence imposes an order on the observations that must be preserved when training models and making predictions.

Generally, prediction problems that involve sequence data are referred to as sequence prediction problems, although there are a suite of problems that differ based on the input and output sequences.

In this tutorial, you will discover the different types of sequence prediction problems.  ... " 

Science of Creativity

AI and Machine learning find us answers, but do they make us curious about deeper values? Discussion podcast:

" ...  Astrophysicist and author Mario Livio discusses his new book on curiosity.

 Curiosity is a fundamental human trait. Everyone is curious, but the object and degree of that curiosity is different depending on the person and the situation. Astrophysicist and author Mario Livio was so curious about curiosity that he wrote a book about it. He recently appeared on the Knowledge@Wharton show on SiriusXM channel 111 to talk about what he learned in the course of writing his book, Why? What Makes Us Curious.

An edited transcript of the conversation follows.

Knowledge@Wharton: What is it that really drives our curiosity?  ... " 

Improving Siris Voice

Looking forward to this change.  Voice can imply intelligence and empathy, and is a very meaningful feature.

How Apple Finally Made Siri Sound mode Human   By David Pierce in Wired

" ... This fall, when iOS 11 hits millions of iPhones and iPads around the world, the new software will give Siri a new voice. It doesn't include many new features or tell better jokes, but you'll notice the difference. Siri now takes more pauses in sentences, elongates syllables right before a pause, and the speech lilts up and down as it speaks. The words sound more fluid and Siri speaks more languages, too. It's nicer to listen to, and to talk to.

Apple spent years re-architecting the technology behind Siri, transforming it from a virtual assistant into the catch-all term for all the artificial intelligence powering your phone. It has relentlessly expanded into new countries and languages (for all its faults, Siri's by far the most worldly assistant on the market). And slowly at first but more quickly now, Apple has worked to make Siri available anywhere and everywhere. Siri now falls under the control of Craig Federighi, Apple's head of software, indicating that Siri's now as important to Apple as iOS.    .... " 

My posts on virtual assistants. 

Advanced Analytics. Interviews with the C Suite

We had much experience in this area, delivering analytics to many executives.   It seems the connections have slipped in general.   It takes more than interviewing them, it takes ongoing connection and review of the goals, processes and the value propositions  involved

Advanced analytics: Nine insights from the C-suite
by Jit Kee Chin, Mikael Hagstroem, Ari Libarikian, and Khaled Rifai

Conversations with hundreds of business leaders reveal nine ways that they are—and are not—adapting to the analytics revolution.

Data and advanced analytics have arrived. The volume of available data is growing exponentially, with more added every day from billions of phones, sensors, payment systems, and cameras. Machine learning is becoming ubiquitous, but organizations are struggling to turn data into value.

The stakes are high. Those who advance furthest, fastest will have a significant competitive advantage; those who fall behind risk becoming irrelevant. Analytics cannot be the sole province of the chief information officer (CIO), as is sometimes the case. The CIO may not understand the business as a whole well enough to spot opportunities and threats, or be influential enough to ensure that the company addresses them appropriately. While the expertise the CIO brings is of course essential, business-unit leaders and CEOs must be in charge of analytics to accelerate the pace of change and to ensure intelligent investment. This is beginning to happen: McKinsey has found that more than 50 percent of CEOs consider themselves the primary leader of the analytics agenda, and that figure has been growing steadily.

With this in mind, we spoke to more than 300 top executives of major companies. Here we offer nine insights based on these conversations, and suggest actions for business leaders to take.  .... " 

Decision Trees in Practice

They can also work because they are transparent.     But that transparency can also show when they are difficult to use operationally.   Which can be a good thing.

Why do Decision Trees Work?   in DSC
Posted by Amelia Matteson 

This article is from Win-Vector LLC: 

In this article we will discuss the machine learning method called “decision trees”, moving quickly over the usual “how decision trees work” and spending time on “why decision trees work.” We will write from a computational learning theory perspective, and hope this helps make both decision trees and computational learning theory more comprehensible. The goal of this article is to set up terminology so we can state in one or two sentences why decision trees tend to work well in practice.  .... " 

Google Express Expanding in Retail

Another strong movement to online retail,  but I do note that many of these interactions are not exclusive.  Many retailers will likely look at multiple options. Who will provide the better and most engaging interaction, to the most people?

Okay Google, I want to order from Home Depot’   By George Anderson in Retailwire. With commentary.

Google is on a retail roll. Home Depot announced it will join Google Express this fall with shoppers being able to place voice orders using Google Assistant with the tech giant’s Home speaker device or mobile app. The Home Depot announcement comes just a couple of weeks after a similar announcement was made by Walmart.

With two of the largest retail chains in the world moving into voice with Google, the question becomes of more a matter of when than if with others selling on the Express platform. Other retailers on Google Express include Bed Bath & Beyond, Costco, Kohl’s, Payless, Pier 1, Toys “R” Us, Ulta, Walgreens and others. .... " 

BBC Launches R&D Unit for Virtual and Augmented Reality

Bottom line,  VR especially needs more content to drive interest from consumers,  AR also needs strong interest from companies looking for strong marketing channels.

BBC launches specific R&D unit for AR and VR research.  By Rebecca Hills-Duty  in VRFocus

The BBC have already taken some steps towards producing content in virtual reality (VR) and 360-degree video. Now that has been solidified with the creation of BBC Reality Labs, a division of BBC R&D focussed on creating content in VR and augmented reality (AR).

BBC Reality Labs will be extending the work the BBC has already done in the VR area and looking into how VR and AR technologies can improve the BBC as a whole. The BBC Reality Labs team plans to work with the W3C web consortium as well as continuing to use the open-source WebVR standard to produce high-quality VR content that is viewable using only a web browser.

Part of the plan for BBC Reality Labs is to further developer WebVR content and examine how web browsers can be used to deliver VR content to audiences.  .... "