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

Thursday, April 27, 2017

Acting Quickly

 In the HBR,     Its part of the risk model.

How to Act Quickly Without Sacrificing Critical Thinking  by Jesse Sostrin

An unbridled urgency can be counterproductive and costly. If you’re too quick to 
react, you can end up with short-sighted decisions or superficial solutions, neglecting underlying causes and create collateral damage in the process.

But if you’re too deliberative and slow to respond, you can get caught flat-footed, potentially missing an opportunity or allowing an emergent challenge to consume you.

To balance these two extremes, you need reflective urgency — the ability to bring conscious, rapid reflection to the priorities of the moment — to align your best thinking with the swiftest course of action. In my work, coaching leaders at every level through a variety of management dilemmas, I’ve developed three strategies to practice reflective urgency: ... "

Transparency not Needed in Health Care Deep Learning?

Not sure I agree.  Will the accuracy and kinds of errors be the same?  Liability implications.  And also, who will be liable for errors that are made?   Will the AMA concur?  That is all wound up in the algorithmic methods. The Neural nets in this case, their architecture and training methods.  In Technology Review: 

Deep Learning Is a Black Box, but Health Care Won’t Mind
New algorithms are able to diagnose disease as accurately as expert physicians.
by Monique Brouillette  

Earlier this year, artificial intelligence scientist Sebastian Thrun and colleagues at Stanford University demonstrated that a “deep learning” algorithm was capable of diagnosing potentially cancerous skin lesions as accurately as a board-certified dermatologist.

The cancer finding, reported in Nature, was part of a stream of reports this year offering an early glimpse into what could be a new era of “diagnosis by software,” in which artificial intelligence aids doctors—or even competes with them.  .... " 

Wednesday, April 26, 2017

Baidu and a Free OS for Cars

Quite an interesting play.  Open source OS for self driving cars.  An unexpected thing from a player like China via Baidu.  Maybe free but not open source?    In Technology Review:

Baidu Will Release a Free Operating System for Self-Driving Cars
China’s leading search engine hopes to speed development of autonomous driving and draw car makers to its services.     by Will Knight  ... 

Amazon Echo Look Fashion Assistant

A kind of 'magic mirror' play.  We experimented with, primarily in cosmetics.  Can that be far away?   Following.   Currently available only by invitation.  In Engadget: 

Amazon Echo Look is a voice-controlled camera for fashion tips
You can show off your wardrobe without taking mirror selfies.
 by Jon Fingas, @jonfingas

Amazon's Echo smart speakers just went in an unusual (but potentially very helpful) new direction. Meet Echo Look, an Alexa-powered camera designed around taking your own fashion photos and videos. If you want to show off your daily wardrobe, you just have to ask the Look to take a snapshot -- you don't have to take a selfie in front of a mirror to get a full-length picture. And since it includes a depth-sensing camera, it can blur the background to make shots pop. The real party tricks come when you're not sure about your outfit, however. ... "

(Update) More detail on Echo Look via Amazon.
(Update)  More detail and invite request.  Looking like much more than Beta.

Bots in the UnCanny Valley

The 'Uncanny Valley'  is a term used in the physical representation of people in android form,  as they get closer to reality we get uneasy. Now it has been suggested that the way bots chat is also reaching that.  I have conversed with bots where they seemed to have a personality.  Even the tone of Alexa when she says something like 'Good Night', can invoke an oddly social but uncanny feeling.  But a longer conversation,  which expects basic human performance, soon breaks that spell. So are we soon likely to combine image features and chat speech?  Will that be more or less disconcerting?

Chatbots Have Entered the Uncanny Valley 
The Atlantic,  Kaveh Waddell

The tendency for people to be repelled by increasingly humanoid and human-like robots may extend to chatbots and digital assistants as well. "The more human-like a system acts, the broader the expectations that people may have for it," says Carnegie Mellon University professor Justine Cassell. Modern chatbots use banter and humor, conversational speech, and parsing free-form questions and answers to coax users into engaging with them in more a human-like manner. "This creates a perception that if you say anything to this bot, it should respond to it," says Autodesk engineer Nikhil Mane. This makes for situations in which user requests exceed the bot's limitations, and the subsequent errors serve to remind users of the assistant's artificiality. Mane says a better approach for bots is to make users aware of their constraints, such as prompting  ... " 

7-Eleven Seeks Japan Labor Efficiency

There are worker shortages in Japan, so 7-Eleven stores are experimenting with RFID tagging to eliminate or greatly simplify checkout, and thus decrease labor needs.    Can this test translate to the US?    Can they tailor the method there and then migrate it.  Can the high cost of RFID be addressed? Recall the tests that Amazon Go is doing with the same idea.  This could a defining US example of retail automation.   More and discussion at Retailwire.

Virtual Assistants Control GE Air Conditioning

Another example of voice control being pushed to common home devices that need control.    A simple example.   Been testing the idea in our home

Don’t get up to turn on your AC unit from GE — just ask Alexa to do it instead   By Lulu Chang

GE Appliances’ connected AC units for windows now integrate with Amazon Alexa for voice-activated capabilities. Thanks to a new Alexa skill from GE Appliances known as Geneva, you can just say, “Alexa, tell Geneva to turn on the air conditioner,” and your air conditioner will obey.

“Consumers today want easier and more convenient ways to control their appliances, and window air conditioners are no exception,” said Mark Evans, product manager of air products at GE. “Now that our connected window air conditioners integrate with Alexa, it’s easier than ever to turn on your air conditioner, adjust the cooling temperature, and more.” ... '

Linkedin Changes its Agreement with You

 A look at how Microsoft is altering the Linkedin user agreement you are very likely to say OK to.  Evan Schuman points out some of the risks.  Linkedin is a place where you are likely to voluntarily offer lots of personal and business information.  How is it protected?

Use LinkedIn a lot? Read its new privacy policy — carefully.
At the very least, it’s a reminder that social media embarrassments are forever  By Evan Schuman

And looking at the scale involved,  Linkedin now has 500M registered users.

Team Stress Levels

In the enterprise saw much lip service given to this kind of issue, but few solutions other than cautions, here an attempt at measure.    In the HBR: 

An Early Warning System for Your Team’s Stress Level
Thomas Hellwig, Caroline Rook, Elizabeth Florent-Treacy, Manfred F. R. Kets de Vries

Tuesday, April 25, 2017

On the Future of Augmented Reality

Israel and Scoble in a Podcast on the near future of Augmented and Virtual reality.  Suggesting that Apple is about to push this strongly.   Via my colleague Walter Riker.

Smarter Manufacturing Robotics with Watson

Interesting play, had never thought of Watson in terms of Robotics.  Note the wide variety of things ABB manufactures, which could lead to efficiency considerations, regarding scheduling and supply chain interactions.

Watson could be the key to smarter manufacturing robots
But will better bots take more jobs?
Rob LeFebvre, @roblef

Some reports predict that robots will replace 5 million jobs in the next couple of years. Bill Gates thinks we need a "robot tax" to compensate for those losses. A new partnership announced today between Swiss automation firm ABB and IBM's Watson initiative could hasten that future while it seeks to improve efficiency in on the manufacturing floor.

Swiss company ABB makes a ton of things, including electric vehicle power, industrial automation systems and manufacturing robots. It's biggest sales, though, come from software that enables communication between machinery and centralized control centers. It's looking to further this growth (and boost its sales) via this new partnership with Watson, IBM's artificial intelligence technology. ... " 

Oracle Wants Advanced Tech

Oracle looking to AI and VR initiatives.

Oracle Launches New In-House 'Startup'
The company wants to stay ahead of the tech game by working on new, cutting-edge initiatives.
  By Julie Bort Business Insider  ... 

Cheap not Deep Learning is the Future

Had not heard of the term 'cheap-learning'  before, but have been thinking of the concept for a long time.  Building higher levels of abstraction is a matter of better leverage.  Only a few people need to know the low level math.  These people are absolutely necessary, but just a few of them.     And very few as we creep towards automation.    A matter of car mechanics vs Phds in thermal mechanics.

In Datanami: 

" ... Higher Levels of Abstraction

In the new cheap learning paradigm, the combination of sophisticated frameworks like Theano and Tensorflow and the powerful but simple languages like Python will help to create a new layer of abstraction that eliminates the need for big data application developers to understand the nitty gritty details of high-level math and low-level execution models to get stuff done.

Developers will be able to tell the computer what to do at a high level, Dunning says, and the computer will take care of the implementation details, whether it’s running on a 1,000-node cluster, a gaggle of GPUs, or just a laptop.

(Ted) Dunning (of MapR)  uses a car analogy to communicate his vision for how the cheap learning metaphor will evolve.

“I don’t understand cars anymore. I understand what combustion is and what gasoline is, but the actual details of how cars work escaped my grasp many years ago,” Dunning says. “But I have a mental model, when I want to get to work or talk to somebody about the car. We need to have that same sort of loss of detail, and a higher abstraction level, when talking about these parallel programs.” .... ' 

Microsoft Channels Process for Intelligence

This sounds remarkably like using/building a process model for given tasks.   We experimented with this do define and decompose jobs-tasks.   Like the idea that these exist and can be re-purposed to make things that look intelligent, by solving sub-tasks.

Microsoft Releases Dialogue Dataset to Make Chatbots Smarter| by Roland Meertens   
Maluuba, a Microsoft company working towards general artificial intelligence, recently released a new open dialogue dataset based on booking a vacation - specifically, finding flights and a hotel.

The number of chat bots has risen recently, especially since Facebook opened their Messenger platform to these bots a year ago. At the moment, most bots only support very simple and sequential interactions. Advanced use cases such as travel planning remain difficult for chatbots. With this dataset Maluuba (recently acquired by Microsoft) helps researchers and developers to make their chatbots smarter.  .... " 

Smart Home Tracks Health

Great thought, but its still a bit of an effort to set things up just to turn the lights on and off.   Our own smart home demonstrations made it clear that most homes had to be wired and outfitted during construction to make this economically viable.  Its coming, but will take infrastructure investments to make it common.   Good example and design visuals in Digital Trends:

This smart house could keep track of your health like a live-in nurse  ..By Garrett Hulfish 

Our health is one of our biggest concerns. This prototype smart home could worry about it for us.
When people get sick, they go to the doctor. However, wouldn’t it be better to keep track of your health on a daily basis? That can take a lot of work, though, and so what if our homes could keep track of it for us? That’s a question that led Joe Colistra and his students to build a prototype smart house unit.

Colistra is the design and planning associate professor at the University of Kansas School of Architecture. Using two awarded grants totaling $51,000, he and his fifth-year students will build part of a home embedded with sensors to collect the biometric data of its residents. .... "

Thoughts on Data Engineering

All analytics is data engineering.  Manipulating data content in context to achieve useful results, using a proven process.

A Data Engineer's Guide To Non-Traditional Data Storage    by Irina Papuc:

Data Engineering
With the rise of  and data science, many engineering roles are being challenged and expanded. ... 

Originally, the purpose of data engineering was the loading of external data sources and the designing of databases (designing and developing pipelines to collect, manipulate, store, and analyze data).
It has since grown to support the volume and complexity of big data. So data engineering now encapsulates a wide range of skills, from web-crawling, data cleansing, distributed computing, and data storage and retrieval.

For data engineering and data engineers, data storage and retrieval is the critical component of the pipeline together with how the data can be used and analyzed. ... "   

Monday, April 24, 2017

IBM Announces Watson Marketing Insights

Just announced, looked at this in beta form.  IBM continues to apply Watson learning to different expertise domains, following retail applications.    Likely to continue see such specialized applications in the future, a reasonable way to start applying cognitive AI methods that augment human approaches improving marketing analysis.  Note the mention of Loyalty in particular.

IBM Introduces New Cognitive-Powered Cloud Offering for Marketers
Watson Marketing Insights Helps Marketing Professionals Drive Customer Loyalty

ARMONK, N.Y., April 23, 2017 /PRNewswire/ -- IBM (NYSE: IBM) today announced IBM Watson Marketing Insights. Available on the cloud, Watson Marketing Insights has cognitive capabilities that can study the behavior of customers and reason how they might impact the success of the business. With this insight, marketers can launch targeted campaigns designed to cultivate customer advocates and enable long-term business success.

IBM Watson Marketing features proven cognitive capabilities that can uncover key predictors in customers by examining their interactions with the company, whether it's in the store or on email, digital and social media. By continuously updating this data, Watson Marketing Insights reveals new audience profiles and customer segments as the relative importance of their behavior predictors changes.  .... " 

Which ML Algorithms to Use?

A somewhat unusual thing to see from SAS.  In the SAS blog, a cheat sheet for using machine learning algorithms for the beginner.   Largely non technical.  Nicely done piece that worth linking to. See also many such cheat sheets by DSC (Data Science Central).  That provide similar approaches. In the link below, a chart that summarizes it all, not complete, but provides a very good overview of the space.  Click through for a larger and expandable chart..

Which machine learning algorithm should I use?  
By Hui Li on Subconscious Musings  

This resource is designed primarily for beginning data scientists or analysts who are interested in identifying and applying machine learning algorithms to address the problems of their interest.

A typical question asked by a beginner, when facing a wide variety of machine learning algorithms, is “which algorithm should I use?” The answer to the question varies depending on many factors, including:

The size, quality, and nature of data.
The available computational time.
The urgency of the task.
What you want to do with the data.

Even an experienced data scientist cannot tell which algorithm will perform the best before trying different algorithms. We are not advocating a one and done approach, but we do hope to provide some guidance on which algorithms to try first depending on some clear factors.   ... " 



Amazon Wants to Handle Your Subscription Replenishment

Pushing the subscription model used to be one of the biggest futurist conversations in CPG enterprise, I am sure the conversation continues.  Assure your sales volume with quality product, and automatic replenishment.  You can make this happen by setting up automatic subscription.  And making it very easy to place that order.  Everything is in place to make this happen.   Amazon holds the reigns.  Seen Amazon Dash?   For most everything we need.  It used to be we were not used to having many kinds of goods shipped to us, but the consumer view of that is rapidly changing.    In Engadget: 

Amazon is trying to be your one-stop subscription shop
'Subscribe with Amazon' is the retailer's latest move to own your shopping. .... "
Timothy J. Seppala, @timseppala  .... 

(Update)   And more in TechCrunch.

AI Driven Clothes Shopping

I think I have seen at least three recommender-engine driven apparel selection systems pitched to me, none have survived.  This is a tough sell.  Touch and 3D would help.  Another mirror like idea utilizing machine vision.   Will AI help this?  I do hope so, the fashion industry could also use the data involved, have seen that from students in my classes.  In Fastcompany, with considerable detail:

The Robots Are Coming For Your Wardrobe

The husband and wife duo behind Chennai-born AI startup Mad Street Den are determined to change the way you shop for clothes.    By Jackie Snow.

Artificial intelligence is rapidly changing just about every industry, with billions being poured into algorithms for everything from self-driving cars to detecting cancer to chatbots. Not so much fashion.

Mad Street Den, a three-year-old artificial intelligence startup founded by a husband and wife team in Chennai, India, is boldly going where few startups have gone before. The couple’s complementary career paths have helped. CEO Ashwini Asokan was previously with Intel Labs working on product design; her husband and CTO, Anand Chandrasekaran, is a neuroscientist who switched over to AI when he realized the technology wasn’t quite there yet to build silicon brains. ... " 

Sensor Filled Gloves

A prototype for gloves that contain health sensors

Sensor-filled glove could help doctors take guesswork out of physical exams

" .... Now, an interdisciplinary team of researchers at the University of California San Diego and Rady Children's Hospital has developed new wearable sensors and robotics technology that could be used to accurately measure muscle stiffness during physical exams. "Our goal is to create a system that could augment existing medical procedures by providing a consistent, objective rating," said Harinath Garudadri, a research scientist at the university's Qualcomm Institute and the project's lead investigator.

"Many clinical exams and procedures are very subjective and rely on measurements that are done with a physician's hands," said Andrew Skalsky, director of the division of Rehabilitation Medicine at Rady Children's Hospital. "We often make major medical decisions and diagnoses based on touch and feel. With this technology, we can start to develop objective measurements for subjective processes." ... ' 

UPS Using Beacons for Accuracy

It sounds like the beacons are being used to get indoor accuracy to affirm location.  And alert the system to misplacement. Reducing error.   In ChainstoreAge: 

UPS uses beacons to quell shipping errors,   By Deena M Amato-McCoy

Preload Smart Scan uses smart technology to avoid this issue. Specifically, the program comprises Bluetooth-enabled beacons that communicate with package-scanning devices worn by UPS employees as they load packages onto vehicles. The scanners, which read package labels, are programed to know where a package belongs in a specific vehicle.  ... "

Sunday, April 23, 2017

Editing Photos with Neural Networks

In NVidia Blog, quite interesting piece on generating new photos from collections of photos.  In other words editing them.   The article is quite technical, but outlines  how new problems are being solved this way.  Also offers some software for your use.  Worth a look for those examining problems and methods.  Note the application of open source TensorFlow.

Photo Editing with Generative Adversarial Networks (Part 1)
Tagged Deep Learning, DIGITS, GAN, TensorFlow

 ... Adversarial training (also called GAN for Generative Adversarial Networks), and the variations that are now being proposed, is the most interesting idea in the last 10 years in ML, in my opinion.
Yann LeCun, 2016 [1].

You heard it from the Deep Learning guru: Generative Adversarial Networks [2] are a very hot topic in Machine Learning. In this post I will explore various ways of using a GAN to create previously unseen images. I provide source code in Tensorflow and a modified version of DIGITS that you are free to use if you wish to try it out yourself  ...." 

A Broad View of Amazon

Interesting view from CBInsights about Amazon.  Pointer to a further strategy document.  I agree, quite a breadth of directions by a former book seller.

Amazon Strategy Teardown: Building New Business Pillars In AI, Next-Gen Logistics, And Enterprise Cloud Apps

Seattle-based Amazon is doubling down on AWS and its AI assistant, Alexa. It's seeking to become the central provider for AI-as-a-service. ... 

Amazon is the exception to nearly every rule in business. Rising from humble beginnings as a Seattle-based internet bookstore, Amazon has grown into a propulsive force in at least five different giant industries: retail, logistics, consumer technology, cloud computing, and most recently, media and entertainment. The company has had its share of missteps — the expensive Fire phone flop comes to mind — but is also rightly known for strokes of strategic genius that have put it ahead of competitors in promising new industries.

This was the case with the launch of cloud business AWS in the mid-2000s, and more recently the surprising consumer hit in the Echo device and its Alexa AI assistant. Today’s Amazon is far more than just an “everything store,” it’s a leader in consumer-facing AI and enterprise cloud services. And its insatiable appetite for new markets mean competitors must always be on guard against its next moves. .... " 

PC Magazine Reviews Watson Analytics

A review in PC Magazine that much likes Watson Analytics, describing it as a self-service analytics tool rather than cognitive.

Saturday, April 22, 2017

Google Building its Own Quantum Chip

Have followed quantum computing from the beginning.  Even offered some industrial test problems for potential solution.  Followed  NASA and Google Tests of research and startup efforts. Good to see real solutions approaching.  Google has the right kind of problems.  See tag below on Quantum efforts.   In Technology Review:

Google’s New Chip Is a Stepping Stone to Quantum Computing Supremacy
The search giant plans to reach a milestone in computing history before the year is out. ...  by Tom Simonite .... " 

Towards an Era of Mass Collaboration

Good overview of the idea, and players that are helping to make it happen.  with similar challenges to any kind of group work measurement.

We're Entering A New Era of Mass Collaboration    By Greg Satell @Digitaltonto  In Inc:

In a networked world, the best way to become a dominant player is to be an indispensable partner .... "

Amazon Rolls Out Lex Chatbot Tools

Amazon rolls out conversational Chatbot interface 'Lex'.  Am looking for better yet conversational and memory capabilities for Lex.   Good technical overview at the link.

Amazon Lex – Build Conversational Voice & Text Interfaces  by Jeff Barr  
While computers that talk are great, computers that listen and respond are even better! If you have used an Amazon Echo, you know how simple, useful, and powerful the Alexa-powered interaction model can be.

Today we are making the same deep learning technologies (ASR – Automatic Speech Recognition NLU – Natural Language Understanding) that power Amazon Alexa available to you for use in your own conversational applications. You can use Amazon Lex to build chatbots and other types of web & mobile applications that support engaging, lifelike interactions. Your bots can provide information, power your application, streamline work activities, or provide a control mechanism for robots, drones, and toys.

Amazon Lex is designed to let you get going quickly. You start out by designing your conversation in the Lex Console, providing Lex with some sample phrases that are used to build a natural language model. Then you publish your Amazon Lex bot and let it process text or voice conversations with your users. Amazon Lex is a fully-managed service so you don’t need to spend time setting up, managing, or scaling any infrastructure.  ... " 

Google Earth is Updated

We were much impressed with Google Earth,  even used it for things like showing where our plants were worldwide, and providing 'tours' to their locations.  Seems not that much has been done to it in some time, but the Verge reports on a recent update.  New 3D capabilities.  Always interesting.

Technology and Societal Change

How Technology Can Help Solve Societal Problems

The Network Revolution

In the next article of the series, “The Network Revolution: Creating Value through Platforms, People and Technology,” authors Barry Libert, Megan Beck, Brian Komar and Josue Estrada debut the concept of Social Change as a Platform. Libert is a Wharton senior fellow and CEO of OpenMatters; Beck is the firm’s chief insights officer. Komar is vice president of marketing for Salesforce.org, the nonprofit reseller of Salesforce.com Inc. Estrada is the senior vice president of strategy and operations at Salesforce.org. 

As Charles Dickens so astutely observed about life during the French Revolution in “A Tale of Two Cities,” it was the best and worst of times. One could say the same thing today. The Fourth Industrial Revolution of technology networks and platforms could usher in an era of mass societal disruption — as well as unprecedented social cooperation. Whether the latter would prevail depends on the ability of nonprofit entities and the broader social sector to boost their collective impact by adopting the new business models that are disrupting the for-profit world. It would also depend on whether they can embrace what we call ‘Social Change as a Platform’ or SCaaP. ... " 

Selfie Drones

Drones that can follow you around to take pictures and perhaps provide security?    In Engadget:

Hover Camera can now follow and record its owner sans phone    Richard Lai, @richardlai ... 

" ... The term "selfie drones" may suggest an added level of convenience, but in reality, these machines still share one common pain point with conventional offerings: they need to be connected to a mobile device for initiation, mode selection and landing. The folks over at Zero Zero Robotics, however, are tackling this issue with a software update for their Hover Camera Passport. As we previewed back at CES, this foldable, fully-enclosed drone is finally getting an "Owner Mode," in which it can follow and record its owner without ever having to turn on the smartphone. ... " 

Friday, April 21, 2017

How Retailers can Cope with Slowing Growth

In  Knowledge@Wharton,  Non -Technical, but quite strategic these days.  Recall talking to Marshall Fisher while in the enterprise.   (See his book)

How Retailers can Cope with Slowing Growth

Wharton's Marshall Fisher discusses why retailers must break their 'addiction' to top-line growth.

Podcast:

"... Successful retailers can grow quickly in their early years simply by opening new stores. But eventually they run out of real estate, and then they need the discipline to stop opening new stores and focus instead on driving more sales through their existing stores. They can boost sales and profits dramatically by making changes in the way they run their existing stores, such as with help from analytics and the use of technology.

In fact, several such small changes brought in profits that helped 17 retailers outperform the stock performance of the S&P 500 index, according to a new study titled “Curing the Addiction to Growth” published in the Harvard Business Review by Marshall Fisher, Wharton professor of operations, information and decisions, along with his co-authors, Vishal Gaur (who has a PhD from the Wharton School and now is a professor at Cornell’s Johnson School) and Herb Kleinberger (who has an MBA from Wharton and for many years led PWC’s retail practice).    .... " 

Mercedes Connecting to Amazon Echo and Google Assistants

The breadth of this is most interesting.  All their new cars can link to both of the most common  assistants.  Seems an indicator of popularity of the assistant idea as yet not seen.   A first true link between Smart Home and Smart Car.    Next steps?  In theVerge: 

Mercedes-Benz is connecting the Amazon Echo and Google Home to all its new cars ...   by Andrew J. Hawkins  @andyjayhawk

Mercedes-Benz announced today that all of its 2016 and 2017 vehicles in the US can now connect with both Amazon and Google’s digital voice assistants.

Starting today, Mercedes owners can instruct their Google Home or Amazon Echo to remotely start or lock their vehicles, as well as send addresses to their in-car navigation system. But a promo video by Mercedes shows a much more frightening use-case: using these digital voice assistants to compensate for incredibly stupid behavior, like leaving the house with both the iron and stovetops on at full blast. ... " 

Related:

Ford is integrating Alexa into its cars.  It was the first to do this.

Hyundai is working with Google Home.

Google Home Recognizes Distinct Voices

A major issue with home assistants, that they cannot distinguish between members of a household, is being remedied this week.  The Google Home assistant, which I have had under test since its inception, can now distinguish up to 6 voices in a household, allowing for different profiles for individual household members.  Which should allow for different, tailored and more intelligent responses. Getting us closer to conversations.   And also be able to capture distinctive household member data.

  In our particular household, that has not been an issue.  Will be testing that.  How accurate? Enough for adult vs child?  Purchasing?  Secure answers?  The Amazon Echo or other assistants do not yet allow for this capability, but rumor is out that it is coming.

(Update) Easy enough to train it for your voice.  But you have to invite others to be trained, and they have to load up their own copy of the 'Google Home' App, to train themselves. .... (More to follow) 

List and links to Assistants studied.  Over 40 assistants have been commented on here.


(Update) Technology review on the issues here.

New Kinds of Assistant Magic Mirrors

We examined the Mirror idea under many conditions.  For cosmetics, eye-ware, clothing.  At home and in the store.  The somewhat new idea then is having trained associates guide its use.  A Virtual assistant.  Does not save the cost of expensively trained labor problem.  Saw a similar thing being done in Asia, where having live demonstration people in the aisle is already common.   But will robotics ultimately take that function? See the Lowes' bot example.

Having a system that can really operate well at home has different challenges,  to make it work and be engaging.   The phone is the natural channel.  We tried permanently mounted mirrors.  Has challenges similar to developing sales chatbots, but with image recognition as a sensory input.   I see this has a limited potential for high end products.   But many will still want to see such products live. Could drive some to stores for a demo.

Mirror, mirror on the wall, will you leave me any shoppers at all?

This magic mirror could be a great sales tool. More likely, though, it will just siphon sales to an online rival.   By Evan Schuman, star Influencer, Contributing Columnist, Computerworld  ... 

The interplay between store associates and in-store technology has always been a delicate balancing act. When the tech helps the associate be an all-knowing partner to the shopper, it's a great thing. But when the tech is deployed so that the associate seems to just get in the way, it can ultimately undermine the in-store experience. Enter the HiMirror.

Retailers have toyed with magic mirrors before — it seems to be the retail tech idea that never dies, despite the fact that it rarely works long term — but this mirror goes much further. ... " 

Why Do Deep AI Methods Work?

Gets back to transparency, and can we actually explain how advanced AI methods work?  This issue was known years ago during early methods with neural network methods.   Our own work in the area was criticized for this.  Does this make a difference?

The Dark Secret at the Heart of AI
No one really knows how the most advanced algorithms do what they do. That could be a problem.

by Will Knight  April 11, 2017

" .... Already, mathematical models are being used to help determine who makes parole, who’s approved for a loan, and who gets hired for a job. If you could get access to these mathematical models, it would be possible to understand their reasoning. But banks, the military, employers, and others are now turning their attention to more complex machine-learning approaches that could make automated decision-making altogether inscrutable. Deep learning, the most common of these approaches, represents a fundamentally different way to program computers. “It is a problem that is already relevant, and it’s going to be much more relevant in the future,” says Tommi Jaakkola, a professor at MIT who works on applications of machine learning. “Whether it’s an investment decision, a medical decision, or maybe a military decision, you don’t want to just rely on a ‘black box’ method.”

There’s already an argument that being able to interrogate an AI system about how it reached its conclusions is a fundamental legal right. Starting in the summer of 2018, the European Union may require that companies be able to give users an explanation for decisions that automated systems reach. This might be impossible, even for systems that seem relatively simple on the surface, such as the apps and websites that use deep learning to serve ads or recommend songs. The computers that run those services have programmed themselves, and they have done it in ways we cannot understand. Even the engineers who build these apps cannot fully explain their behavior. .... " 

Whats Interesting in Your Data?

In MapR they pose the question: Cognitive Analytics Answers the question: What's Interesting in Your Data?

Good general thought.  But I would suggest it depends what your definition of interesting is. Measurably valuable in current and continuing business process and practice?   Yes.

That is why its cognitive.   People and process are embedded in the question and measures provided by analytics.

Singapore Plans its Future City

BBC takes an overview look.   Very impressive plans, but there needs to be much more detail.  For example how will security be handled? Privacy of data?   Lots of challenges here, but this is very likely the future.

Tomorrow's Cities: Singapore's plans for a smart nation   By Karishma Vaswani,   Asia business correspondent ... "

Thursday, April 20, 2017

Social Media Demographics

A view of most major social media their demographics and buying habits in 2017.  Interesting, but overly infographic.  Could be done in a single visual page.   Useful information.

Bain on the Firm of the Future

Not so much technology oriented, but about the basic notion of the firm.  Good, broad view.   Especially for technologists, who often believe it is all about tech advances.  Its also about business process and how to deliver value.

The Firm of the Future

April 12, 2017 Bain Brief By James Allen, James Root and Andrew Schwedel

In the world of the firm, something is changing. It’s not that your local bookstore went out of business. Or that your taxi driver now rates you on a 5-point scale. Or that anything can now be outsourced, allowing even the smallest firms to rent capabilities on demand. It’s more profound than these.

The prevailing paradigm that has underpinned business for the past 50 years is under review. The simplest version of that paradigm is that firms exist first and foremost to deliver returns to their shareholders’ capital—and the sooner they deliver it, the better. We will describe the challenges confronting this paradigm. But the first question we asked as we observed the changes was this: Is such a shift unusual? Has the idea of the firm been consistent over time, or has it changed before? ... " 

GE Digital Approach to Digital Transformation

Doing work with GE Digital on an upcoming Analytics Summit.

They send along this Webinar, reviewing their approaches:

" ... Gain insights into the five pillars of digital industrial transformation based on GE’s own playbook. .... "

How Viruses Spread

An impressive visual and description of of how viruses spread.  In Quartz   " ... A new genetic tool maps how deadly viruses spread around the world in real time ... ".   Recall a recent post on Kellogg work in using cell data for prediction.  Tag below on Epidemics.   And also links to our work on retail data for the same purpose.

Robotics Newsletter

Konrad Holubek writes:
 
" ... I wanted to let you know that I collect and send out a weekly email on robotics, a subject I am very passionate about. 

You can see an example of the latest newsletter here : http://snip.ly/v4kjf

If you are also interested in robotics, you can subscribe to the newsletter here : http://immersive-robotics.com/subscribenewsletter.html

Please note that the newsletter is spam free : no ads, just carefully curated news about robotics.  ... " 
 

Retail Analytics Trends

Fairly general look at retail analytics and their future.

In Bridge2i
Categories Analytics, Blogs Customer Intelligence, Retail
Retail Analytics Trends 2017 and Beyond .... 

Introduction to Machine Learning

A good, largely non technical guide to the basics of machine learning.  Useful for the interested and motivated decision maker or executive.   Part I here, others apparently to follow.

Wednesday, April 19, 2017

Procter Singapore Supply Chain Center

In Business Journal, about Procter&Gamble:

" ... The Singapore E-Center will be the Cincinnati-based company’s (NYSE: PG) first of its kind outside the U.S. and is designed to fuel end-to-end digital innovation across its supply chain management, e-analytics and e-business by expanding partnerships with local small-medium enterprises and startups and supporting their development of digital solutions.... "

Microsoft Expands Cognitive Services

I remain disappointed by the integration level of Cortana assistant with other MS services, but at least now Microsoft seems to be advancing on cognitive services.  Now developers can take the capabilities further.  Includes mention of their 'Build a Bot Strategy'.  But very little about Cortana itself, except as a front end.  Seems they are losing an AI opportunity here to Amazon and Google. Need a framework for assistants, and Microsoft has one of the most popular in Office.

Microsoft makes three more cognitive services generally available
Microsoft is rolling out more of its Cognitive Services programming interfaces, which provide developers with a way to add face recognition, computer vision and more to their apps and services.
By Mary Jo Foley  .... " 

End of Brand Loyalty?

Of much importance to my former employer and all of CPG.  Still don't see its involved with culture. just the channels that cultures currently use for information.   Those channels can be selected by marketing.  In Forbes: 

The Death Of Brand Loyalty: Cultural Shifts Mean It's Gone Forever   by Kathleen Kusek

Oracle and Moat for Measurement and Analytics

Its all about accurate and universal measurement, directly and indirectly through the results of analytics.  So this is not surprising.

Oracle buys Moat
Creates the World's Most Comprehensive Cloud Platform for Marketing Data and Analytics
REDWOOD SHORES, Calif., April 18, 2017 /PRNewswire/ -- Oracle today announced that it has signed an agreement to acquire Moat, the fastest-growing digital measurement cloud company. Moat will join Oracle Data Cloud, which uses data and analytics to enhance media for leading marketers and publishers. The combination of Oracle and Moat will connect data to consumer attention for better media experiences and business outcomes. .... " 

Facebook using Augmented Reality

Zuckerberg talking how AR will play in Facebook.    And suggests that it will first play through the cameras in people's phones.   Less cumbersome than a headset,  but less stability too.  In VRFocus:

The Camera To Be ‘The First Augmented Reality Platform’ – Zuckerberg
 Facebook will use the camera part of the Facebook app as the basis for AR platform. ... " 

Tuesday, April 18, 2017

Talk: Artificial Intelligence in the Car

Upcoming CSIG Talk this week.  

" ... This a reminder about our Cognitive Systems Institute Speaker Series on Thursday, April 20, 2016 at 10:30 am ET US (7:30 am PT).  Our speakers are Roberto Sicconi and Maggie Styś, Founders of TeleLingo, who will be presenting "Artificial Intelligence in the Car." ... 

Please point your web browser to https://apps.na.collabserv.com/meetings/join?id=2894-8491 password=cognitive.   Use audio on computer or dial 855-233-7153  PIN Code: 43179788    Non-IBMers please use the "guest" option instead of entering your email.   

 ...  A link to slides and a recording of each call should be available on the CSIG website (http://cognitive-science.info/community/weekly-update/).   We encourage those who join the calls to add questions and comments to the LinkedIn Discussion Group:  https://www.linkedin.com/groups/Cognitive-Systems-Institute-6729452  .... 

Thank you!
Dianne Fodell  ....   " 

Microsoft Flow and Buttons

In SiliconAngle: 

" ... (Microsoft) The company has also introduced something it calls Flow-compatible hardware buttons. They’re somewhat reminiscent of Amazon.com Inc.’s Dash buttons, which allow consumers to restock their supplies at the push of a button in their pantry, for example. ... 

However, Microsoft is aiming at the Internet of Things with this new update. It’s partnered with The Button Corporation and Shortcut Labs, so that businesses can install web-connected Bttns from Button Corp. in conference rooms and other locations. This should help employees to gain help when problems crop up, without needing to create a ticket or make a phone call, Microsoft said. .... " 

Using AR To Transfer Images

If you can't draw well, or precisely, perhaps you can trace images from a phone App using Augmented Reality.   Clever idea for the right kind of application.  Warns that this is best done with the aid of a tripod.  Available on IOS for now,   Right now it only works with A4 size paper, but they report that it will ultimately be linked to the Project Tango system to detect location,  on a Lenovo Phone, with the potential of doing full size murals.

In ITunes:  " ... SketchAR is an application through which the user sees a virtual image on the surface of which they are planning to trace a sketch. In one hand, the user holds a phone and sees a virtual image, while the other hand traces virtual lines already on the paper ... " 

SketchAR puts virtual images on paper to let you trace drawings from your phone by Dami Lee @dami_lee In TheVerge: .. 

Simplified Heart Rate Sensor Tech

In Engadget.   Another interesting breakthrough in health sensor technology.  Possible use for other emotion sensing applications.

Panasonic can read your heart rate by looking at your face
The sensor tech will be used to show the stress levels of golfers at an upcoming tournament. ... "

How to Mislead with Data

 I have seen it often in the enterprise and elsewhere.  Good points made in article below, and always worth repeating.    But missing is the current tendency to dress the results. Aka 'Tell a Story', or use some sort of pre-established narrative.  Or use an infographic to simplify into an instantly understandable point to be made.    Yes, that's confirmation bias, or model-confirmation bias.

Also very common and hard to detect unless you are shown the raw data and can manipulate it, which is rare.  Helps to do a risk analysis to understand the cost of a wrong model. Or involve the context owner early and often.

How to Lie with Data  Posted by Karolis Urbonas    In DSC

We expect that data scientists and analysts should be objective and base their conclusions on data. Now while the name of the job implies that “data” is the fundamental material that is used to do their jobs, it is not impossible to lie with it. Quite the opposite – the data scientist is affected by unconscious biases, peer pressure, urgency, and if that’s not enough – there are inherent risks in the process of data analysis and interpretation that lead to lying. It happens all the time while the intentions might be truly honest – though we all know the saying “The road to Hell is paved with good intentions”.

As every industry in every country is affected by data revolution we need to make sure we are aware of the dangerous mechanisms that can affect the output of any data project..... " 

Monday, April 17, 2017

DeepMind Open-Sources Sonnet

DeepMind, now part of Alphabet, has open sourced Sonnet, a TensorFlow library for constructing neural network modules that allows variable sharing.  Another win for broader experiments with machine learning.

About Deepmind, now part of Alphabet.

Their open sourcing announcement:  https://deepmind.com/blog/open-sourcing-sonnet/   Indicates they plan to continue to update the resources.

Technical resources in Github.

(I corrected the title and other lines, my error, too much terminology)
 

Microsoft Updates Flow and PowerApps

Been waiting for more about this.   This interests me because it addresses workflow and thus business process.  IFTTT does that as well, but in a somewhat restricted approach.  Still instructive.  Looking to dig deeper here in understanding how they compare.  Azure linkages.  We need more succinct ways to model business process based on flows of contextual data.

Microsoft launches updates to PowerApps and Flow    by Frederic Lardinois in TechCrunch:

" ... Microsoft’s business-centric IFTTT competitor Flow and its ‘low-code’ PowerApps platform are both getting major updates today. While these are obviously different services that solve different issues, both aim to help non-developers make better use of their existing data and services without having to write their own code. While Flow, at its core, uses the if-this-then-that model, PowerApps allows you to build your own desktop and mobile applications from scratch.

Out of those two services, Flow is still the more approachable one and with today’s update, Microsoft is adding a number of new features that make it more useful for building integrations that streamline the kind of approval workflows that are still the bane of many enterprise users. To help users build these approval processes, Flow is getting new templates and a streamlined design experience for them in this update (though it looks like this feature may not go live before the end of the week).... " 

More on Microsoft PowerApps.

More on Flow in SiliconAngle.


Shrinking Deep Learning Neural Networks

Via O'Reilly.  As these tasks take on tougher problems,  efficient architecture will be important.   Even more important as these tasks become real-time.

Excerpt video:
Shrinking and accelerating deep neural networks    Song Han on compression techniques and inference engines to optimize deep learning in production. ... "

Learning by Sketching with CogSketch

Quite interesting idea.  I remember looking at the CogSketch platform.  Had not of thought of it as a learning method.  But many of us do learn visually, can this be a means to fill the learning gaps?   Does it work in contexts other than geology, that are less visual?   How do the analogy aspect work?

Helping Students Learn by Sketching
Sketch Worksheets software analyzes and provides feedback on student sketches

Northwestern University professor Ken Forbus and his team have developed Sketch Worksheets, software that helps students learn via sketching exercises and also provides on-the-spot feedback by analyzing sketches and comparing them to the instructor's drawings. The software is founded on CogSketch, an artificial intelligence platform previously developed in Forbus' lab that employs visual-processing algorithms to automatically replicate and understand human-drawn sketches. 

Sketch Worksheets' comparisons of student and instructor sketches are conducted by an analogy model, in which students and instructors apply conceptual labels to their sketches to represent relationships among the drawings' different components. Forbus says CogSketch uses analogy to compare labels and give feedback. Geoscientists at the University of Wisconsin-Madison used Sketch Worksheets to devise a set of 26 sketches that cover topics in introductory classes. "This is a step in creating software that can communicate with people as flexibly as we communicate with each other," Forbus says. .... " 

See the tag below for more research information regarding Ken Forbus, previously mentioned.

Resolving Conflict in Self-Managed Teams

Further, perhaps more important, when should a team be self managed?   In the HBR:

How Self-Managed Teams Can Resolve Conflict   by Amit Maimon

In a traditional team structure, conflicts can be escalated to the boss to resolve. Can’t agree on how to prioritize projects, or on which deadlines need to shift? Ask the team leader to step in and make a call. Think a coworker is acting snarky, or that their work is too sloppy? Advise the manager to give them some feedback. But for flat or self-managed teams, that’s not an option. Self-managed teams must identify different ways to find and address day-to-day conflicts.

Self-managed teams can focus on three things to help them successfully resolve conflicts. (Traditionally hierarchical teams may benefit from them too.)   .... " 

Chrome Secure by Default?

Recently pointed to this in the Google blog.  Which claims that Chrome is secure by default.  Is this a reasonable claim?   Am a long time Chrome browser user.  Even used a Chromebook for a test.   Have seen no reason to move to Edge.   Have seen, rarely,  a few of the 'warnings' when entering a URL, indicating that the target destination was 'a deceptive site'.  How complete is that kind of warning?  How well does the classification of a 'deceptive site' work?   Quite an interesting data science problem by itself.

Competing Neural Nets Getting Smarter

We experimented with a similar ideas, using simulation and evolutionary methods to rate alternate neural network designs.  Considerable improvements since then.  Net architecture still contains much art.  The 'art' we talk about here can be readily rated by performance.  In Wired:

Google’s Dueling Neural Networks Spar to Get Smarter, No Humans Required

THE DAY RICHARD Feynman died, the blackboard in his classroom read: “What I cannot create, I do not understand.”

When Ian Goodfellow explains the research he’s doing at Google Brain, the central artificial intelligence lab at the internet’s most powerful company, he points to this aphorism from the iconic physicist, Caltech professor, and best-selling author. But Goodfellow isn’t referring to himself—or any other human being inside Google. He’s talking about the machines: “What an AI cannot create, it does not understand.”

Goodfellow is among the world’s most important AI researchers, and after a brief stint at OpenAI—the Google Brain competitor bootstrapped by Elon Musk and Sam Altman—he has returned to Google, building a new research group that explores “generative models.” These are systems that create photos, sounds, and other representations of the real world. Nodding to Feynman, Goodfellow describes this effort as an important path to all sorts of artificial intelligence. .... " 

Sunday, April 16, 2017

BBC Looks at Robots and Workers

Rise of the robots: What advances mean for workers  By Tim Harford .... No details here, but an interesting set of examples from the BBC.  

Count Down to Jambaar Time

Jambaar.com:  The Video-as-a-Service Platform for Data & Technology-Enabled Business Growth   (4 Days from Today)    @JambaarCom

An Immersive Star Wars Land

Been reading Brian Jay Jones biography of George Lucas: A Life.  I am  a long time fan of digital cinema and further how it can be made more immersive, so seeing that evolution in our lifetime is particularly interesting.  It is an amazing story of how Lucas brought about Star Wars, and now its translation to an immersive experience by Disney.

 In The Verge, the latest in that journey.
Watch the creation of Disney’s ambitious and immersive Star Wars land   “Everything about it is just about total immersion.”  by Bryan Bishop@bcbishop   Is the future one of immersion rather than dark space projection?   Expected to open in Florida and CA in 2019.   More background in Wikipeda.

Obituary: Robert Taylor, Computing Pioneer

I never met Computing Innovator Robert Taylor, but through connections at the Pentagon, ARPA and PARC heard of his innovative management that led to advanced computing and networking.  Notable too for his work on the Alto, progenitor of the personal computer,  and interaction with Douglas Engelbart on innovative computing interfaces.   ... In the NYTimes:

Robert Taylor, Innovator Who Shaped Modern Computing, Dies at 85
By John Markoff ...

In the Wikipedia.

Saturday, April 15, 2017

Virtualitics: Immersing Yourself in Data

Have been taking a look at Virtualitics, a means to immerse yourself with and collaborate with complex data using VR.  Clever idea we experimented with before VR was easier to do.

" ... In SiliconAngle: Based out of Pasadena, California, Virtualitics offers a data visualization platform that combines artificial intelligence, Big Data and virtual reality to create interactive 3D data models. Virtualitics’ platform is a VR application that allows users to generate complex data models composed of up to 10 dimensions of data, all in a shared virtual office where users can collaborate and analyze the data together. Using machine learning, Virtualitics’ platform is also able to recommend optimal visualizations that gives users a better understanding of the whole picture for their data. ... "

Robots and Surveillance Teams

Researchers link robots into surveillance team   By Bill Steele

Rod Kurcoba, Cornell Engineering Communications

Seeing many possible views of the same area from fixed and mobile cameras could be confusing to a human, but a computer can combine it all, track people and objects and notice significant events.  

If you were monitoring a security camera and saw someone set down a backpack and walk away, you might pay special attention – especially if you had been alerted to watch that particular person. According to Cornell researchers, this might be a job robots could do better than humans, by communicating at the speed of light and sharing images.... " 

Passive RFID Detects Cracks

A kind of sensor we examined for a packaging quality control application, continues to evolve.

Passive RFID Tracks Changes in Structural Micro-Cracks

Dai Nippon Printing has developed a passive adhesive sheet with a built-in Identiv UHF RFID tag, to detect when a crack to which a sheet is affixed widens, and to transmit that event when interrogated.
By Claire Swedberg ..." 

Amazon Go

Has been a while since I have heard much about Amazon Go, which is a very limited test to create no-checkout retail.   This was one of the top of mind approaches discussed in our innovation centers. No doubt this will ultimately become common.  Still lots of details to work out, but it can ultimately be a win. A recent look in e-commerce Times:   Will Amazon Go Reinvent Retail? 

An Omni Channel Data-Driven Retailer

Interesting retail from Ronald van Loon     ...  Red through for the suggestions.

How to Become an Omni-Channel Data-Driven Retailer

Everyone's Blog Posts -  DSC. by Ronald van Loon 

In today’s digital age where customers are as likely to buy a product from an eCommerce website as from a brick and mortar store, delivering a seamless and value-adding shopping experience has become more important than ever before.|

 The multiple shopping channels available to customers and the competition posed by other retailers have made it an absolute necessity for a retail business to integrate data inputs from different channels and use to it define an omni-channel shopping experience. 

As a Big Data and BI influencer, I have worked with a large number of businesses within and beyond the retail industry. Using this knowledge and expertise, I have developed a 5-step approach that businesses operating in the retail sector can adopt to meet the expectations of their customers and become an omni-channel, data-driven retailer.   .... "

Friday, April 14, 2017

Proposed, An AI Index to Track Automation

See some of our previous look at Job and task analysis.  See the O*Net tag below, and the associated  online resource here.  A place to start for this kind of analysis.  But it is still hard to get precise breakouts of jobs to tasks, and how they are likely to change in the presence of emerging technology. Try it first with past data on automation with standard computing.

New Tools Needed to Track Technology's Impact on Jobs, Panel Says
By The New York Times  via the CACM.

New tools must be developed to track and measure the impact of artificial intelligence (AI) and automation on the U.S. job market, according to a report from an expert panel at the National Academies of Sciences, Engineering, and Medicine.

The panel recommends creating an AI index, analogous to the Consumer Price Index, which would be integrated with data on skills and tasks involved in various occupations to guide education and job-training programs. It also says a public-private alliance is necessary for developing such tools because information from numerous sources, ranging from traditional government statistics to databases from online services, can be harnessed for insights on skills, job openings, and how effective training programs are. The idea is that employees in declining occupations could find useful information about more promising careers requiring both similar and new skills. The tool also could pull information on job placement rates for courses that cultivate those new skills. ....  "

Collaborative Machine Learning Training

In CACM.   A kind of crowd sourcing of training.

Google Tests New Approach to Training Machine Learning Models  in eWeek  
Google is testing a new collaborative machine-learning training method in which training data is diffused across millions of individual mobile devices instead of housed in datasets distributed across servers in the cloud. Google says machine-learning models can be trained from user interaction with their Android devices using the Federated Learning strategy.

"Federated Learning enables mobile phones to collaboratively learn a shared prediction model while keeping all the training data on [the] device," say Google's Daniel Ramage and Brendan McMahan. They note this decouples "the ability to do machine learning from the need to store the data in the cloud."   

Predicting Epidemics (Bioterror?) from Cell Data

If you have followed this blog for a long time, we looked at how to predict epidemics through OTC drug purchase data.  Here is a related idea that might be combined with earlier work.  Links to Bioterror detection?  See the tag links below.

Using Cell Phone Data to Predict the Next Epidemic    Insight Kellogg.
Who you call is linked to where you travel, which dictates how viruses spread.

Based on the research of Pierre Deville, Chaoming Song, Nathan Eagle, Albert-László Barabási and Dashun Wang  ... Can big data about whom we call be used to predict how a viral epidemic will spread?  .... " 

Uber and Behavioral Economics

Fascinating piece about behavioral economics.      Behavioral economics seems to be OK when everyone agrees that it is good and fair, by their definition.  Back to the nudge, with some basic guidelines:   Nudge me only for clearly aligned company and employee interests, with a clear permission to opt out.    Or is it more an environment than a nudge?

Uber Shows How Not to Apply Behavioral Economics  by Francesca Gino
A recent New York Times article on how Uber is using various insights from behavioral economics to push, or nudge, its drivers to pick up more fares — sometimes with little benefit to them — has generated quite a bit of criticism of Uber. It’s just one of several stories of late that have cast the company in a poor light.

When I read the piece, it reminded me of a question executives often ask me when I talk to them about the benefits of behavioral economics or give them examples of how they could use it in their own organizations: “Aren’t you afraid it will be used with ill intent?”  ... '

Closer Look at Facebook M

Technology Review takes a close and an insightful look at the Facebook M assistant.   Long awaited and in broad test.  Though I am not a fan of Facebook in general.   I like the idea that new assistants, and  'skills' written for them, will give us new understanding of the current limitations of delivering AI.

And as was revealed long ago, Facebook M integrates human help into the mix.  Apparently seamlessly.   That's not bad, since ultimately we will all be working with these assistants.  Back to the word concierge.

I also think that an assistant can be very successful by just being reasonably focused on some value.  And entertainment/music can be that value.  Opening it up for later leverage in other areas.  Its certainly not about just doing  'smart' tricks.    Looking to get my hands on Facebook M.

Facebook’s Perfect, Impossible Chatbot
Facebook is quietly trying to develop the most useful virtual assistant ever, in a project that illustrates the current limitations of artificial intelligence.   by Tom Simonite  April 14, 2017

Amazon’s Alexa can summon an Uber and satisfy a four-year-old’s demand for fart noises. Siri can control your Internet-connected thermostat. Each serve millions of users each day. But a lucky group of around 10,000 people, mostly in California, know that Facebook’s assistant, named M, is the smartest of the bunch.  .... " 

My link index of all assistants examined.

Thursday, April 13, 2017

Face to Face Better than an EMail

Not sure of the 34x difference,  quite a bit of context involved.  But some motivation as to why face to face is better.

A Face-to-Face Request Is 34 Times More Successful than an Email
Vanessa K. Bohns  in the HBR .... 

Semantic Document Processing

Today's call was of interest, and showed how complex document understanding can be:

" ... Today ... Our speaker was Sridhar Iyengar, IBM Distinguished Engineer at the IBM T. J. Watson Research Center, who will be presenting "Semantic PDF Processing & Document Representation."  ....  Slides here.   and Recording here

Please find the schedule of presenters herefor the next several calls.   A link to slides and a recording of each call should be available on the CSIG website (http://cognitive-science.info/community/weekly-update/).   We encourage those who join the calls to add questions and comments to the LinkedIn Discussion Group https://www.linkedin.com/groups/Cognitive-Systems-Institute-6729452  .... 

Thank you!
Dianne Fodell

Augmented Reality Guidelines

Not exactly standards, and more details would be very useful, following up

New guidelines point to an augmented future
Lockheed Martin, Procter & Gamble and Caterpillar joined 62 other organizations to create hardware and software guidelines for using augmented reality on the manufacturing floor ....

(Update) With more details in IndustryWeek.
Which mentions the Augmented Reality for Enterprise Alliance (AREA)
See also the AREA blog which contains some interesting example.

Beyond, to a VR Facemask

Reading emotions was one of our early efforts to understand consumer interactions with products.   Called Expression Glasses developed at the Affective Computing Project at MIT.  This seems to take it yet beyond that.   For what applications?   In the Verge:

This VR face mask can read your emotions  by Adi Robertson
" ... The new MindMaze Mask, which was announced today, is a ring of electrodes that can be installed in any VR headset’s foam face mask. When you put the headset on, the mask detects which sensors your skin is touching with a certain pressure, then matches the pattern to one of 10 facial expressions, which an avatar reproduces. There’s also some software-based prediction, which is supposed to reduce lag between your face and your avatar’s. It’s not detecting and mirroring every movement — it’s more like you’re pressing button combinations with your face to call up animations. The example avatars are simple cartoons, but they can look like anything a developer wants. ,,,"

P&G and Amex, Building AI Past and Present

Tom Davenport mentioned some of our past AI work at P&G in his recent HBR article.  Below an introductory excerpt, much more detail at the link.   While considerable improvements have been made in available technology since then, some of the same key challenges exist.

How can advanced technical methods (Logic-based Expert Reasoning back then,  Deep Learning neural nets now)  be integrated with business process and applied to provide intelligent, but adaptive reasoning to real business systems?  Then how can these systems be effectively tested, maintained and reapplied in new contexts.  How can decision makers and the consumer understand their implications and risks? Many of us are working on that now.

How P&G and American Express Are Approaching AI
Thomas H. Davenport and Randy Bean

" ... P&G also built and employed a number of rule-based expert systems. In addition to Peri, the current CDO, we also spoke with Franz Dill, a retired P&G IT manager who focused on AI during the 80s and 90s. He said that the most well-known expert system they developed was one that blended Folgers coffee (no longer a P&G brand). This system, Dill noted, saved P&G in excess of $20 million dollars a year in green coffee costs. The company also built an expert system that helped advertisers at P&G to use, modify, and reuse the company’s advertising assets.

Both American Express and P&G are companies that have explored artificial intelligence over the years, and while the technology may have changed, the established yet innovative approaches that these firms take to incorporating new technologies and capabilities continues to evolve. Their fundamentally sound innovation practices provide a foundation for evolution. ... " 

Wednesday, April 12, 2017

Examples of AI and Machine Learning in Industry

From O'Reilly:  Some intriguing examples.

Free: AI and Machine Learning in Industry

In Artificial Intelligence and Machine Learning in Industry: Perspectives from Leading Practitioners, you'll hear from Michael Osborne on coming effects of automation, Arjun Singh on AI in education, Jake Heller on AI and the law, Aaron Kimball on "intelligent microbes" (and their commercial use), Bryce Meredig on how to apply machine learning for inventing new materials, and Erik Andrejko on machine learning in agriculture.

And it's free with registration, a 38 page document, compliments of Microsoft. Download your copy here  .... 

Manufacturing Needs Digital Ready Infrastructure

In the Cisco Blog.  Makes sense.  Note the mention of VR, which led me to this in a search for practical examples.

Why Do Factories Need a Digital-Ready Infrastructure?   by Nada MacKinney

Digital transformation, digital business, digitization,  digital disruption. There are countless ways to phrase it, but it boils down to this. When digital technologies enable new business models, that’s digital transformation. Digital factory initiatives are generating incredible business outcomes, and at the core there’s a common denominator—a digital-ready infrastructure.

Manufacturers’ Distinct Take on Digital Transformation

In manufacturing, digital transformation means consumers using virtual reality to design their own products. It includes collaboration across multiple engineering teams or between customers and plant floor experts. It means robotics bringing new levels of automation and accuracy to the plant floor. Digital transformation cranks up efficiency by connecting every extremity of the supply chain for total visibility. It pushes efficiency by using sensors to track assets, tools and people for improved safety and security. Digitally savvy manufacturers increase agility by connecting machines for predictive maintenance, resulting in fewer unplanned  ...  " 

Creating a Happy Workplace

Podcast: On creating a happy workplace.  In Knowledge@Wharton.   A study done with firefighters.  Perhaps an unusual context to map against the typical office?  Or perhaps not.

Virtual Reality in Merchandising

In Retailwire, with a discussion.

Will virtual reality transform in-store merchandising?

hrough a special arrangement, presented here for discussion is a summary of a current article from the monthly e-zine, CPGmatters.

Virtual reality (VR) simulation is arguably the most talked-about tech solution in grocery today. It aims to drive faster, smarter and more profitable decisions by enabling retailers to ideate new in-store concepts, evaluate them with test shoppers, and activate those concepts.

The technology is “a good value,” Tom Wozbut, director, category analytic and shelving hub, Kellogg Co., said in a presentation he did at the Western Michigan University 2017 Food Marketing Conference last month along with Jason Smith, VP, customer service, InContext, its VR technology provider. ..... 

Structured Problem Solving

And taking it beyond to business process models?    Not commonly done well enough to define the structure.

In Sloan Review: 

Saving Money Through Structured Problem-Solving

Research Feature March 21, 2017  Reading Time: 6 min
Nelson P. Repenning, Don Kieffer, and Michael Morales

Closely observing how work really gets done in your organization can yield numerous opportunities for improvements.

Executives can’t lead effective organizational change just by sitting in their offices. As busy as they are, leaders need to find ways to observe fundamental work processes in their organizations. When they do, they usually discover that there are gaps between concept and reality in how work gets done. Michael Morales’ experience — in which identifying and addressing such gaps led to his company saving $50,000 in just 60 days — is a case in point. .... " 

Tuesday, April 11, 2017

Drawing for Everyone

This is always something I wanted, something that turned me into a graphic artist.   A Clever AI experiment by Google:

Fast Drawing for Everyone
Dan Motzenbecker,  Creative Technolohist, Google Creative Lab

 ... Drawing on your phone or computer can be slow and difficult—so we created AutoDraw, a new web-based tool that pairs machine learning with drawings created by talented artists to help you draw. ... It works on your phone, computer, or tablet (and it’s free!). So the next time you want to make a birthday card, party invite or just doodle on your phone, it’ll be as easy and fast as everything else on the web. ....   " 

Infographics and Data Science Visual Resources

DSC does its usually good job of pulling together lots of interesting visual resources.    " ... They come in various shapes: infographics, cheat sheets, periodic tables, one-picture articles, and even maps. They cover everything: IoT, AI, machine learning, data science, deep learning, Hadoop, Python, R, dataviz, statistical theory, big data - you name it. Below is a selection that is most relevant to our readers. ...  " ...  by Vincent Granville

Electronic License Plates for Drones

In IEEE Spectrum

Electronic License Plates for Drones
In late 2015, mandatory drone registration went into effect in the United States. Since then, anyone who wants to fly a drone (or model aircraft) weighing over 0.55 pounds (0.25 kilograms) must register with the U.S. Federal Aviation Administration to receive a unique identification number. This number needs to be placed on the drone, but there is no requirement for it to broadcast signals to allow for remote identification. That might change in the future, as the FAA considers different mechanisms to identify drone owners and operators. Last month, DJI, the world's largest commercial drone manufacturer, announced a proposal outlining a general scheme for doing just that. ... " 

CPG Companies and Data Assets

Good to see this push, complex and deep data assets point towards the need for better analytics.  The idea of data as an asset has been much mentioned here, and experimented with.  See the Data Assets tag below.

Playing catch-up in advanced analytics
By Luis Benavides, Rehana Khanam, Frédéric Lefort, and Oscar Lovera-Perez

Consumer-packaged-goods (CPG) companies have increasingly gained access to unprecedented amounts of data, and we expect that this trend will intensify over time. However, most companies are generating very limited insight from these newly found sources, leaving a treasure chest of opportunities untapped. Companies that have already captured benefits from this trove of insight are reaching the next frontier of performance. For those companies that have yet to follow suit, it is imperative to take steps to catch up. ... " 

In McKinsey: Among CPG companies’ most underused assets are the vast quantities of data they generate. But there’s still opportunity to invest in industry-beating analytic capabilities. .... " 

Via McKinsey Insights.

More Assistant Competitors to Alexa

Rumors out about assistant oriented competitors to Amazon Echo.  The latest being a collaboration between Harman-Kardon and Microsoft.  In Neowin: " .. .Harman Kardon is working on a Cortana-powered speaker for a 2017 release ... "    By Rich Woods @TheRichWoods  ...

More previously in Neowin on the idea.  The idea makes sense, use the speaker/entertainment idea to drive adoption, then use Cortana as a way to lead the idea to an assistant or to a smart home hub.

Microsoft has already been talking about a home hub announcement this year.   This may be the year to see this competition expand.   In another piece, we see that Sonos says it will integrate Alexa voice smarts into its new offerings.   The question remains, beyond voice control, what kind of actual intelligence will be seen in these offerings?   Will it be enough to sell the devices beyond convenient entertainment?

Saving Shopping Malls by Tracking?

We looked to this kind of tracking for better personalization,  but not much has happened in the last decade.

Will customer tracking help save shopping malls?    by Matthew Stern In Retailwire: 

With shopping malls not performing like they used to, facility owners are looking to technology to revive shopper interest. But when it comes to re-mapping the mall to make it more attractive to shoppers, some of the technologies they are using might put some customers on edge.

Some mall landlords have been using smartphone monitoring technology to track shopper behavior throughout the malls they own, according to The Wall Street Journal. Some monitor the amount of time customers spend in the mall and in specific stores to determine where one store should be set up in proximity to another. Landlords also base advertising on location data used in conjunction with social media and email information.

While the Journal article doesn’t detail which specific solutions are used for each example of behavior tracking, it does mention one big U.S. mall owner that uses a tool called StepsAway to push discounts and promotions to shoppers’ smartphones. Shoppers in this case do not need to install an app.   ... " 

Monday, April 10, 2017

Captioning Images with Tensorflow

 Below, Nicely done and instructive piece on captioning using neural methods.   Captioning as a means of multi tagging of images.  Now this we could have used when captioning copy images and documents for retrieval, analysis and re-use.   How accurate in the given context?  How much editing might be required?   In O'Reilly:

Caption this, with TensorFlow
How to build and train an image caption generator using a TensorFlow notebook.  By Raul Puri, Daniel Ricciardelli  March 28, 2017

Image caption generation models combine recent advances in computer vision and machine translation to produce realistic image captions using neural networks. Neural image caption models are trained to maximize the likelihood of producing a caption given an input image, and can be used to generate novel image descriptions. For example, the following are possible captions generated using a neural image caption generator trained on the MS COCO data set.     ... "