By 

JONATHAN VANIAN

 

October 2, 2018

Google and Facebook are teaming up to make each company’s artificial intelligence technologies work better together.

 

The two companies said Tuesday that an unspecified number of engineers are collaborating to make Facebook’s open source machine learning PyTorch framework work with Google’s custom computer chips for machine learning, dubbed Tensor Processing Units, or TPU. The collaboration marks one of the rare instances of the technology rivals working together on joint tech projects.

 

“Today, we’re pleased to announce that engineers on Google’s TPU team are actively collaborating with core PyTorch developers to connect PyTorch to Cloud TPUs,” Google Cloud director of product management Rajen Sheth wrote in a blog post. “The long-term goal is to enable everyone to enjoy the simplicity and flexibility of PyTorch while benefiting from the performance, scalability, and cost-efficiency of Cloud TPUs.”

Facebook product manager for artificial intelligence Joseph Spisak said in a separate blog post that “Engineers on Google’s Cloud TPU team are in active collaboration with our PyTorch team to enable support for PyTorch 1.0 models on this custom hardware.”

 

Google first debuted its TPUs in 2016 during its annual developer conference, and pitched them as a more efficient way for companies and researchers to power their machine-learning software projects. The search giant sells access to its TPUs via its cloud computing business instead of selling the chips individually to customers like Nvidia, whose graphics processing units, or GPUs, are popular with researchers working on deep learning projects.

 

Artificial intelligence technologies like deep learning have grown in popularity over the years with tech giants like Google and Facebook that use the technologies to create software applications that can automatically do tasks like recognize images in photos.

As more businesses explore machine learning technology, companies like Google, Facebook, and others have created their own AI software frameworks, essentially coding tools, intended to make it easier for developers to create their own machine-learning powered software. These companies have also offered these AI frameworks for free in an open source model in order to popularize them with coders.

Read and Learn More

 

 

 

 AI is massively transforming our world, but there’s one thing it cannot do: love. In a visionary talk, computer scientist Kai-Fu Lee details how the US and China are driving a deep learning revolution — and shares a blueprint for how humans can thrive in the age of AI by harnessing compassion and creativity. “AI is serendipity,” Lee says. “It is here to liberate us from routine jobs, and it is here to remind us what it is that makes us human.”

Learn More

Depression is a leading mental disorder impacting about 16 million Americans. According to the World Health Organization
Depression is a leading mental disorder impacting about 16 million Americans. According to the World Health Organization

Depression is a leading mental disorder impacting about 16 million Americans. According to the World Health Organization, the annual global economic impact of depression is estimated at $1 trillion and is projected to be the leading cause of disability by 2020.

As researchers aim to better predict, diagnose and treat depression, artificial intelligence is being explored as a potential solution.

Some of the questions that need answering to better understand the role of artificial intelligence in efforts to diagnose and treat depression:

  • What types of AI applications are currently in use to manage depression?
  • How has the market responded to these AI applications?
  • Are there any common trends among these innovation efforts – and how could these trends possibly contribute to reducing the rates of people living with depression?

Depression AI Applications Overview

The majority of AI use-cases for managing depression appear to fall into three major categories:

  • Virtual Counseling: Companies are developing software using machine learning to recognize episodes of depression and to provide support using natural language processing.
  • Patient Monitoring: Machine learning is employed to monitor patients and to predict and prevent the onset of a mental health crisis.
  • Precision Therapy: Firms are using machine learning analytics to track and correlate cognitive function, clinical symptoms and brain activity.

Virtual Counseling Using Natural Language Processing

Woebot

San Francisco-based startup Woebot said its chatbot uses machine learning and natural language processing to help users manage their mood and mitigate depression.

Accessed through the Facebook Messenger platform, Woebot prompts users with questions to assess their mood. Over time, the algorithm, which is trained on cognitive behavior therapy (CBT) methods, learns the emotional profile of each user and recommends activities to help maintain a more balanced mood.

The early-stage startup bases its clinical significance on the results of a clinical trialconducted in partnership with Stanford University. The study demonstrated that Woebot users aged 18-28 experienced “significant reductions in anxiety and depression” compared with the control group that used an e-book published by the National Institutes of Health (NIH).

Woebot was used every day or almost daily by 85 percent of participants for two weeks. Effectiveness was measured using a standard patient health questionnaire for depression called PHQ-9 with scores ranging from 0 (no symptoms) to more than 20 for severe depression.

The chatbot hit 50,000 users in its first week and receives roughly 1 million messages each week and has secured Series A funding. However, the amount of funding is not specified. Woebot, which was launched in June 2017, is currently available for free but a paid version may be in the works as the startup moves toward a sustainable business model.  

Read More

Facebook and AI

Social media accounts are an extremely popular method for staying connected and for using emerging technology to grow audiences. Almost 2.5 billion users worldwide will be present on at least one type of social media channel by the end of 2018.

It’s no wonder that marketing companies as a whole are shifting their focus toward social media and even using AI systems to give themselves an advantage. As the number of tools available starts to rapidly increase it could be possible for marketers to have an easier and more targeted journey of discovering their users.

Facebook and AI:

Facebook with one of the first social networks to start focusing in on AI in the year 2013 when they started to work with New York University. AI systems are present all over Facebook, and many deep learning algorithms are working within keyword studies, shared likes and more. The unfortunate side of Facebook is that it got into hot water from the research that he did with Cambridge analytica. With this specific idea, AI was working to solve through a series of problems over the network. The only issue with that is the AI systems and data that were presented were delivered without the consent of any user. Facebook and the network itself was stopped short in its efforts.

AI and Pinterest:

With so many people on Pinterest often ending up making purchases of the projects that they see online, the app began to use AI learning to recommend products based off of their likes. The system itself had a stylistic nature that ensured that the suggestions presented could be streamlined for users online and designed in such a way that Pinterest could continually present new ideas and make suggestions for users.

This system did not fall under the same type of scandal that was present on networks like Facebook. The Pinterest recommendations are widely welcomed in many cases for the streamlined and stylistic nature of the social network. The way that searching habits have been changed have also made for a social network that is more intelligent because of its suggestions.

Twitter and AI:

Twitter also has an advanced AI built into its current configuration. The AI on Twitter is directly responsible for establishing a category for every single tweet. The idea of the category system is to make sure that the content people care about most is what’s appearing in their timeline.

Twitter is also using a series of neural networks to automatically cropped photos and produce a more consistent aesthetic for the fee. Machine learning was done using eye tracking records to make sure that photographs can be presented in their most appealing form to each user.

Read More

What's the Purpose of Companies in the Age of AI?
Recent advances in artificial intelligence (AI) and computer technology are causing us to think again about some really basic questions: what is a firm? What can firms do better than markets? And what are the distinctive qualities of firms in a world of smart contracts and AI?

While there has been a lot of discussion about “what’s left for humans?” as AI improves at exponential rates — the customary answer is that humans need to focus on the things they are uniquely good at, such as creativity, intuition, and personal empathy — I think we now have to ask, “what’s left for firms?”

In many ways this is an old question, because it takes us back to the arguments of Nobel Laureates Ronald Coase and Oliver Williamson that firms exist to coordinate complex forms of economic activity in an efficient way.  If computer technology has the capacity to simplify and streamline transaction costs, more and more work can be done through these smart-contract arrangements, making traditional human-managed firms obsolete.  For example, when you say to Alexa “order more dog food,” a chain of activities is initiated that leads to the delivery of a fresh supply of Kibble 24 hours later, with little or no human intervention. This work is coordinated by a single firm, Amazon, but it often involves third parties (makers of dog food, delivery companies) whose systems interact seamlessly with Amazon’s.

But is this coordination logic, this ability to internalize transactions to make them more efficient, really the raison d’etre of firms? I would argue that it is just one among many reasons that firms exist. And as computer technology simplifies and reduces transaction costs further, it is these other things that firms do uniquely well that will come more to the forefront. Here are four areas where firms excel.

1. Firms create value by managing tensions between competing priorities. 

In today’s parlance, firms have to exploit their established sources of advantage (to make profits today) while also exploring for new sources of advantage (to ensure their long-term viability).  However, getting the right balance between these two sets of activities is tricky because each one is to a large degree self-reinforcing. Hence the notion of organizational ambidexterity — the capacity to balance exploitation and exploration.

Artificial intelligence is evidently helping many firms to exploit their existing sources of advantage — whether through process automation, improved problem-solving or quality assurance.  Artificial intelligence can also be useful in exploring new sources of advantage: in the famous case of AlphaGo, the winning “strategy” was one that no human player had ever come up with; and computers are increasingly writing new musical scores and painting Picasso-like landscapes.

But AI is not helpful in managing the tension between these activities, i.e. knowing when to do more of one or the other.  Such choices require careful judgment — weighing up qualitative and quantitative factors, being sensitive to context, or bringing emotional or intuitive factors into play. These are the capabilities that lie at the heart of organizational ambidexterity and I don’t believe AI can help us with them at all right now. IBM’s recently-announced Project Debater is a case in point: it showed just how far AI has come in terms of constructing and articulating a point of view, but equally how much better humans are at balancing different points of view.

2. Firms create value by taking a long-term perspective.

As a variant of the first point, firms don’t just manage trade-offs between exploitation and exploration on a day to day basis, they also manage trade-offs over time. My former colleagues Sumantra Ghoshal and Peter Moran wrote a landmark paper arguing that, unlike markets, firms deliberately take resources away from their short-term best use, in order to give themselves the chance to create even more value over the long term.  This “one step back, two steps forward” logic manifests itself in many ways — risky R&D projects, pursuing sustainability goals, paying above-market wages to improve loyalty, and so on. We actually take it for granted that firms will do many of these things, but again they involve judgments that AI is ill equipped to help us with.  AI can devise seemingly-cunning strategies that look prescient (remember AlphaGo) but only when the rules of the game are pre-determined and stable.

An example:  the “Innovator’s Dilemma” is that by the time it’s clear an invasive technology is going to disrupt an incumbent firm’s business model, it’s too late to respond effectively. The incumbent therefore needs to invest in the invasive technology before it is definitively needed. Successful firms, in other words, need to be prepared to commit to new technologies in periods of ambiguity, and to have a “willingness to be misunderstood,” in Jeff Bezos’s terms.  This isn’t an easy concept for AI to get used to.

3. Firms create value through purpose — a moral or spiritual call to action.

There is a second dimension to long-term thinking, and that is its impact on individual and team motivation. We typically use the term purpose here, to describe what Ratan Tata calls a “moral or spiritual call to action” that leads people to put in discretionary effort — to work long hours, and to bring their passion and creativity to the workplace.

This notion that a firm has a social quality — a purpose or identity — that goes beyond its economic raison d’etre is well established in the literature, from March and Simon through to Kogut and Zander.  But it still arouses suspicion among those who think of the firm as a nexus of contracts, and who believe that people are motivated largely through extrinsic rewards.

My view is that you just need to look at charities, open source software movements, and many other not-for-profit organizations to realize that many people actually work harder when money is not involved. And it is the capacity of a leader to articulate a sense of purpose, in a way that creates emotional resonance with followers, that is uniquely human.

Successful firms, in other words, institutionalize a sense of identity and purpose that attracts employees and customers. Ironically, even though blockchain technology is — by definition — about building a system that cannot be hacked, or misused by a few opportunists, people still prefer to put their faith in other people.

4. Firms create value by nurturing “unreasonable” behavior.

There are many famous cases of mavericks who succeeded by challenging the rules, such as Steve Jobs, Elon Musk, and Richard Branson. With apologies to George Bernard Shaw, I think of these people as unreasonable — they seek to adapt the world to their view, rather than learn to fit in. And if we want to see progress, to move beyond what is already known and proven, we need more of these types of people in our firms.

Unreasonableness is antithetical to the world of AI. Computers work either through sophisticated algorithms or by inference from prior data, and in both cases the capacity to make an entirely out-of-the-box leap doesn’t exist. Consider the case of investment management, where robo advisors are not just making trades, they are also providing investment advice to investors, and at a fraction of the cost of human financial advisors. But as the Financial Times said last year, “when it comes to investing, human stupidity beats AI.”  In other words, if you want to beat the market, you need to be a contrarian — you need to make investments that go against the perceived wisdom at the time, and you need to accept the risk that your judgment or your timing might be wrong.  Both qualities that — at the moment — are distinctively human.

So one of the distinctive qualities of firms is that they nurture this type of unreasonable behavior.  Of course, many firms do their best to drive out variance, by using tight control systems and punishing failure. My argument is that as AI becomes more influential, though the automation of basic activities and simple contracts, it becomes even more important for firms to push in the other direction — to nurture unorthodox thinking, encourage experimentation, and tolerate failure.

In a recent Fast Company article, Vitalik Buterin described how all the elements of Uber’s ride-sharing service could be provided through Ethereum-based applications that worked seamlessly with one another:  “the whole process is basically as before, but without the middleman [Uber].”  This is may be true, but it doesn’t necessarily follow that a computer-mediated service is the better option.

Read More

IBM claims that Summit is currently the world’s “most powerful and smartest scientific supercomputer” with a peak performance of a whopping 200,000 trillion calculations per second.
IBM claims that Summit is currently the world’s “most powerful and smartest scientific supercomputer” with a peak performance of a whopping 200,000 trillion calculations per second.

IBM and the U.S. Department of Energy’s Oak Ridge National Laboratory (ORNL) today unveiled Summit, the department’s newest supercomputer. IBM claims that Summit is currently the world’s “most powerful and smartest scientific supercomputer” with a peak performance of a whopping 200,000 trillion calculations per second. That performance should put it comfortably at the top of the Top 500 supercomputer ranking when the new list is published later this month. That would also mark the first time since 2012 that a U.S.-based supercomputer holds the top spot on that list.

Summit, which has been in the works for a few years now, features 4,608 compute servers with two 22-core IBM Power9 chips and six Nvidia Tesla V100 GPUs each. In total, the system also features over 10 petabytes of memory. Given the presence of the Nvidia GPUs, it’s no surprise that the system is meant to be used for machine learning and deep learning applications, as well as the usual high performance computing workloads for research in energy and advanced materials that you would expect to happen at Oak Ridge.

IBM was the general contractor for Summit and the company collaborated with Nvidia, RedHat and InfiniBand networking specialists Mellanox on delivering the new machine.

“Summit’s AI-optimized hardware also gives researchers an incredible platform for analyzing massive datasets and creating intelligent software to accelerate the pace of discovery,” said Jeff Nichols, ORNL associate laboratory director for computing and computational sciences, in today’s announcement.

Read More

Artificial Intelligence can make healthcare more accessible, affordable
Photo: iStockphoto

Most of us approach healthcare with hesitation. There are often several deterrents to seeking medical care (let alone preventive care)—ease of access being a major one. According to the Indian Journal of Public Health (September 2017 edition), India had just 4.8 practicing doctors per 10,000 population. While this is expected to grow to 6.9 doctors per 10,000 people by 2030, the minimum doctor-to-patient ratio recommended by the World Health Organization (WHO) is 1:1000.

Can we have more doctors? That is easier said than done. Even if we can, with our education system focused on quantity over quality, churning out doctors in droves will not guarantee better quality medical services in our country. Besides, with most graduates preferring lucrative urban locations, many Indians still find themselves at great physical or economic distance from quality healthcare.

Essentially, what we need is to fill the gap between the needs of the plenty and the services of the few. In my view, artificial intelligence (AI) has the capability to enable solutions that form the critical middle layer of access—making healthcare accessible and affordable to a large population base at the same quality level irrespective of people’s social standing.

Let us see how AI is set to impact our health in the coming years.

Reactive versus proactive healthcare

The usual attitude towards healthcare is, “I’ll cross that bridge when I come to it”—a reactive rather than proactive approach to seeking medical intervention. However, that has started changing in recent times. Sensors in wearables such as smart watches and Fitbits are already equipped to deliver actionable feedback to apps in our phones and connect to our doctor’s clinic for diagnostic tests and medication prescriptions.

Today, for instance, Apple watches can detect a variety of heart diseases—including diabetes prediction with an 85% match in known cases. All this by using simple, non-invasive tech that is already available.

Interesting AI start-ups like SigTuple (digitized blood analysis), Niramai (thermal scans for breast cancer) and Ten3T (portable, easy-to-use electrocardiograms) are currently developing disruptive diagnostic solutions that will significantly bring down costs, while making physical distance a non-issue. They (and others in this area) do this via cloud-based linkages to hospitals and clinics, chatbots, smart apps and AI-enabled data analytics. This means that sensors, real-time tracking and analytics will enable us to take pre-emptive charge of our own health, helping us to live more aware, healthier, longer lives.

Eliminating human biases

Barring simple ailments, most health consultations and treatments today come with some human bias. Sometimes, there are doubts if doctors’ or pharma companies’ vested interests are pushing certain treatments and medicines. That is why we gravitate towards known doctors. And for serious illnesses and critical care, second opinions are always recommended. In this context, AI-enabled medical care can save time, effort and costs through easy access to unbiased, consistent, good-quality diagnosis and treatment.

The experience of doctors also determines the options they explore. AI makes it possible to access the learnings and data from hundreds of thousands of cases. Oncology, for example, is an area where doctors are continually combining and recombining drugs to treat cancers or overcome resistance to previously successful drugs. Already, AI algorithms are helping doctors analyse a much wider scope of data and predicting—with greater granularity—new drug combinations that are personalized for a patient’s specific need.

Democratizing healthcare for 1.3 billion

Since technology can provide the middle layer bridge, AI-led systems have the potential to take healthcare to people irrespective of their location and affordability. People who live in rural or far-flung locations no longer have to be deprived of the up-to-date care offered at the nerve-centres of medical research. The tech increases the accessibility manifold. We thus have an opportunity to make this world truly equal.

Read More

AI,Chatbot,CMS,CRM,Customer Experience Management,Customer Journey,CX, Digital Experience,IoT,Machine Learning,Marketing Automation,Predictive Advertising

Customer experience is what sets you apart from your competition. A lot of dollars are being invested to analyze customers’ expectations and building technology that can enhance how customers perceive your brand.

There is a lot of action, but there are five technological innovations that are expected to take CX to the next level.
 

1. Cybernetic CX
 

AI is at the core of Cybernetic CX – a cyclic process – analyzing, identifying problems, determining solutions, applying them, monitoring, and repeat.

Cybernetic CX will use advanced analytics and AI to detect patterns and identify anomalies. This information will be fed to machine learning algorithms, which will continue to evolve and be able to correlate with a set of outliers to the root cause. As the problems are diagnosed and the remedies applied, machine learning algorithms powered by heuristics will be able to correctly predict remedies, which will be automatically applied to fix the problem.

Better still, it may even anticipate an upcoming issue and take actions to mitigate it.

According to a J.D.Power study, American Express has excelled at customer satisfaction for their credit cards. They seem to be getting their cybernetics right. For instance, a customer doesn’t have to go through multiple hand-offs while connecting to AMEX departments, as their routing system uses advanced analytics, predictive modeling and operational consolidation to route to the correct department.

Though in an embryonic stage, cybernetics can augment your future CX efforts.
 

2. Digital experience platform
 

For weaving a seamless customer journey, a customer-centric view and integration of all activities like marketing, sales, operations, customer service etc. have become a mandate. DXPs (digital experience platform) help you centralize and share context and content across your organization, which enables ease of coordination and knowledge sharing across locations, teams and technology platforms.

These platforms generally are armed with analytics, CRM, CMS, and marketing automation tools that help them to:

  • Capture customer information and generate a 360-degree view of a customer
  • Keep track of customer lifecycle for service reps to provide a seamless and consistent experience
  • Glean insights to optimize marketing programs
  • Use analytics to understand customer pain points and attribute CX to marketing efforts and in effect optimize marketing programs
  • Help create personalized content across channels
     

Also Read: What to do and not-to-do with chatbots
 

3. Voice of Customer Program
 

Technological solutions like social media monitoring, cross-channel surveys, speech and text analytics are used to capture and analyze customer preferences, feedback, and expectations. VoC tools can give insights that can aid frontline agents to understand their customers better and help various departments (marketing, sales etc.) to have an in-depth view of the customer journey. VoC helps in:

  • Formulating better campaign messages
  • Creating a unified customer view
  • Uncovering areas of opportunity
  • Identifying areas of customer dissatisfaction
  • Measuring business efficiency and performance

A Voc tool may contain:

  • Ability to collect a large amount of customer feedback and generate reports
  • A holistic view of customer journey like the type of interaction, touchpoints etc.
  • NLP, text analytics, speech recognition, semantic analysis, emotion detection etc. 

4. AR/VR

Augmented and Virtual Reality(AR/VR) are the game changers when it comes to creating awesome customer experiences. Both AR and VR can create engaging customer-brand interactions.

AR is being adopted by retail, financial, healthcare and hospitality industries alike to create immersive and meaningful experiences. For instance, AR in the food and beverage industry enhance guest experiences. AR menus create virtual food with multiple digital renderings and 3D photographs to display accurate representation and portions. Customers can also scan menus or food packages to determine nutritional information.

VR too is increasingly complementing CX because of its life-like experiences and emotional footprint.
 

5. IoT

Internet of Things is how various devices form a wireless network and communicate with each other using sensors. IoT holds tremendous engaging power and is the key to bring coherence to omnichannel CX strategies. Leveraging IoT, businesses can:

  • Reach customers in real-time: As a loyal customer is nearing your store, using hisgeo-location you can offer to serve him his favorite meal or offer a discount on his favorite order.
  • Make lives convenient: How about reading a grocery list on your customer’s smartphone and automatically creating a cart with the discounted items, and sending an alert to her to hit the buy button, before she runs out of stock?
  • Product health: IoT product can report its health to the customer care, which can proactively act by scheduling a service and fix issues before they become a reality. Read More

Elon Musk has stated his multi billion-dollar crusade to stop ai space travel

Elon Musk has stated his opinion that AI could lead to the extinction of humanity, and it’s one of the reasons he’s working hard to make us a multi-planetary species. Stephen Hawking was incredibly clear as well: true AI could be the “worst thing” for humanity.

And yet, every country and major company is racing to build AI systems.

Small wonder: Russian president Vladimir Putin has said that the nation that leads in AI will be the ruler of the world. And China is investing heavily in winning the race.

I was in Moscow recently to speak at Skolkovo Robotics Forum. One of the highlights: a visit to Russia’s top cybernetics institute, the National University of Science and Technology, or MISiS.

I asked two of its leaders about AI, its dangers, and — of course — one of the tasks we might use AI for: self-driving cars. Read More