Daily Archives: March 21, 2017

Artificial intelligence (AI) research has explored a variety of problems and approaches since its inception, but for the last 20 years or so has been focused on the problems surrounding the construction of intelligent agents systems that perceive and act in some environment. In this context, intelligence is related to statistical and economic notions of rationality colloquially, the ability to make good decisions, plans, or inferences. The adoption of probabilistic and decision-theoretic representations and statistical learning methods has led to a large degree of integration and cross-fertilization among AI, machine learning, statistics, control theory, neuroscience, and other fields. The establishment of shared theoretical frameworks, combined with the availability of data and processing power, has yielded remarkable successes in various component tasks such as speech recognition, image classification, autonomous vehicles, machine translation, legged locomotion, and question-answering systems.

As capabilities in these areas and others cross the threshold from laboratory research to economically valuable technologies, a virtuous cycle takes hold whereby even small improvements in performance are worth large sums of money, prompting greater investments in research. There is now a broad consensus that AI research is progressing steadily, and that its impact on society is likely to increase. The potential benefits are huge, since everything that civilization has to offer is a product of human intelligence; we cannot predict what we might achieve when this intelligence is magnified by the tools AI may provide, but the eradication of disease and poverty are not unfathomable. Because of the great potential of AI, it is important to research how to reap its benefits while avoiding potential pitfalls.

The progress in AI research makes it timely to focus research not only on making AI more capable, but also on maximizing the societal benefit of AI. Such considerations motivated the AAAI 2008-09 Presidential Panel on Long-Term AI Futures and other projects on AI impacts, and constitute a significant expansion of the field of AI itself, which up to now has focused largely on techniques that are neutral with respect to purpose. We recommend expanded research aimed at ensuring that increasingly capable AI systems are robust and beneficial: our AI systems must do what we want them to do. The attached research priorities document gives many examples of such research directions that can help maximize the societal benefit of AI. This research is by necessity interdisciplinary, because it involves both society and AI. It ranges from economics, law and philosophy to computer security, formal methods and, of course, various branches of AI itself.

In summary, we believe that research on how to make AI systems robust and beneficial is both important and timely, and that there are concrete research directions that can be pursued today.

If you have questions about this letter, please contact Max Tegmark.

To date, the open letter has been signed by over 8,000 people. The list of signatories includes:

Stuart Russell, Berkeley, Professor of Computer Science, director of the Center for Intelligent Systems, and co-author of the standard textbook Artificial Intelligence: a Modern Approach. Tom Dietterich, Oregon State, President of AAAI, Professor and Director of Intelligent Systems Eric Horvitz, Microsoft research director, ex AAAI president, co-chair of the AAAI presidential panel on long-term AI futures Bart Selman, Cornell, Professor of Computer Science, co-chair of the AAAI presidential panel on long-term AI futures Francesca Rossi, Padova & Harvard, Professor of Computer Science, IJCAI President and Co-chair of AAAI committee on impact of AI and Ethical Issues Demis Hassabis, co-founder of DeepMind Shane Legg, co-founder of DeepMind Mustafa Suleyman, co-founder of DeepMind Dileep George, co-founder of Vicarious Scott Phoenix, co-founder of Vicarious Yann LeCun, head of Facebooks Artificial Intelligence Laboratory Geoffrey Hinton, University of Toronto and Google Inc. Yoshua Bengio, Universit de Montral Peter Norvig, Director of research at Google and co-author of the standard textbook Artificial Intelligence: a Modern Approach Oren Etzioni, CEO of Allen Inst. for AI Guruduth Banavar, VP, Cognitive Computing, IBM Research Michael Wooldridge, Oxford, Head of Dept. of Computer Science, Chair of European Coordinating Committee for Artificial Intelligence Leslie Pack Kaelbling, MIT, Professor of Computer Science and Engineering, founder of the Journal of Machine Learning Research Tom Mitchell, CMU, former President of AAAI, chair of Machine Learning Department Toby Walsh, Univ. of New South Wales & NICTA, Professor of AI and President of the AI Access Foundation Murray Shanahan, Imperial College, Professor of Cognitive Robotics Michael Osborne, Oxford, Associate Professor of Machine Learning David Parkes, Harvard, Professor of Computer Science Laurent Orseau, Google DeepMind Ilya Sutskever, Google, AI researcher Blaise Aguera y Arcas, Google, AI researcher Joscha Bach, MIT, AI researcher Bill Hibbard, Madison, AI researcher Steve Omohundro, AI researcher Ben Goertzel, OpenCog Foundation Richard Mallah, Cambridge Semantics, Director of Advanced Analytics, AI researcher Alexander Wissner-Gross, Harvard, Fellow at the Institute for Applied Computational Science Adrian Weller, Cambridge, AI researcher Jacob Steinhardt, Stanford, AI Ph.D. student Nick Hay, Berkeley, AI Ph.D. student Jaan Tallinn, co-founder of Skype, CSER and FLI Elon Musk, SpaceX, Tesla Motors Steve Wozniak, co-founder of Apple Luke Nosek, Founders Fund Aaron VanDevender, Founders FundErik Brynjolfsson, MIT, Professor at and director of MIT Initiative on the Digital Economy Margaret Boden, U. Sussex, Professor of Cognitive Science Martin Rees, Cambridge, Professor Emeritus of Cosmology and Astrophysics, Gruber & Crafoord laureate Huw Price, Cambridge, Bertrand Russell Professor of Philosophy Nick Bostrom, Oxford, Professor of Philosophy, Director of Future of Humanity Institute (Oxford Martin School) Stephen Hawking, Director of research at the Department of Applied Mathematics and Theoretical Physics at Cambridge, 2012 Fundamental Physics Prize laureate for his work on quantum gravity Luke Muehlhauser, Executive Director of Machine Intelligence Research Institute (MIRI) Eliezer Yudkowsky, MIRI researcher, co-founder of MIRI (then known as SIAI) Katja Grace, MIRI researcher Benja Fallenstein, MIRI researcher Nate Soares, MIRI researcher Paul Christiano, Berkeley, Computer Science graduate student Anders Sandberg, Oxford, Future of Humanity Institute researcher (Oxford Martin School) Daniel Dewey, Oxford, Future of Humanity Institute researcher (Oxford Martin School) Stuart Armstrong, Oxford, Future of Humanity Institute researcher (Oxford Martin School) Toby Ord, Oxford, Future of Humanity Institute researcher (Oxford Martin School), Founder of Giving What We Can Neil Jacobstein, Singularity University Dominik Grewe, Google DeepMind Roman V. Yampolskiy, University of Louisville Vincent C. Mller, ACT/Anatolia College Amnon H Eden, University Essex Henry Kautz, University of Rochester Boris Debic, Google, Chief History Officer Kevin Leyton-Brown, University of British Columbia, Professor of Computer Science Trevor Back, Google DeepMind Moshe Vardi, Rice University, editor-in-chief of Communications of the ACM Peter Sincak, prof. TU Kosice, Slovakia Tom Schaul, Google DeepMind Grady Booch, IBM Fellow Alan Mackworth, Professor of Computer Science, University of British Columbia. Ex AAAI President Andrew Davison, Professor of Robot Vision, Director of the Dyson Robotics Lab at Imperial College London Daniel Weld, WRF / TJ Cable Professor of Computer Science & Engineering, University of Washington Michael Witbrock, Cycorp Inc & AI4Good.org Stephen L. Reed, ai-coin.com Thomas Stone, Co-founder of PredictionIO Dan Roth, University of Illinois, Editor in Chief of The Journal of AI Research (JAIR) Babak Hodjat, Sentient Technologies Vincent Vanhoucke, Google, AI researcher Itamar Arel, Stanford University, Prof. of Computer Science Ramon Lopez de Mantaras, Director of the Artificial Intelligence Research Institute, Spanish National Research Council Antoine Blondeau, Sentient Technologies George Dvorsky, Contributing Editor, io9; Chair of the Board, Institute for Ethics and Emerging Technologies George Church, Harvard & MIT Klaus-Dieter Althoff, University of Hildesheim, Professor of Artificial Intelligence; Head of Competence Center Case-Based Reasoning, German Research Center for Artificial Intelligence, Kaiserslautern; Editor-in-Chief German Journal on Artificial Intelligence Christopher Bishop, Distinguished Scientist, Microsoft Research Jen-Hsun Huang, NVIDIA CEO John Schulman, UC Berkeley & OpenAI Koichi Takahashi, PI at RIKEN, Co-chair of Whole Brain Architecture Initiative, CIO of Robotic Biology Institute JT Turner, Knexus Research Corp Vernor Vinge, San Diego, Professor Emeritus of Computer Science Steve Crossan, Google Charina Choi, Google Matthew Putman, CEO of Nanotronics Imaging Owain Evans, MIT, Ph.D. student in probabilistic computing Viktoriya Krakovna, Harvard, Statistics Ph.D. student, FLI co-founder Janos Kramar, FLI researcher Ryan Calo, U. Washington, Assistant Professor of Law Heather Roff Perkins, U. Denver, visiting professor Tomaso Poggio, Director, Center for Brains, Minds and Machines Joshua Greene, Harvard, Associate Professor of Psychology Anthony Aguirre, Santa Cruz, Professor of Physics, co-founder of FLI Frank Wilczek, MIT, Professor of Physics, Nobel Laureate for his work on the strong nuclear force Marin Soljacic, MIT, Professor of Physics, McArthur Fellow, Founder of WiTricity Max Tegmark, MIT, Professor of Physics, co-founder of FLI and FQXi Meia Chita-Tegmark, Boston University, co-founder of FLI Michael Vassar, founder of MetaMed and ex-president of MIRI (then known as SIAI) Sen Higeartaigh, University of Cambridge, Executive Director, CSER Andrew Snyder-Beattie, Oxford, Future of Humanity Institute Project Manager (Oxford Martin School) Cecilia Tilli, Oxford, Future of Humanity Institute researcher (Oxford Martin School) Geoff Anders, founder of Leverage Research JB Straubel, co-founder of Tesla Sam Harris, Project Reason Ajay Agrawal, U. Toronto James Manyika, McKinsey James Moor, Dartmouth Wendell Wallach, Yale Sean Legassick, MobGeo Shamil Chandaria, London U, Institute of Philosophy Michele Reilly, Turing Inc. Michael Andregg, Fathom Computing Ulrich Junker, IBM Miroslaw Truszczynski, University of Kentucky Christian Steinruecken, University of Cambridge, graduate student in AI Mark Waser, Digital Wisdom Institute Douglas Clark, CEO, Mtier Steven Schmatz, University of Michigan Corey Henderson, Computer Security Researcher Jeffrey D. Rupp Amit Kumar, VP & GM, Yahoo Small Business Jesus Cepeda, PhD in Robotics and AI, Monterrey, Mexico Rodolfo Rosini, CEO, Storybricks CD Athuraliya, Machine learning student, USJP, WSO2 Kathryn McElroy, UX Designer for IBM Watson Massimo Di Pierro, DePaul University Anirban Bhattacharya, Computer Science Researcher Lan Laucirica, SpaceX Jesse Brown, UC San Francisco, Neuroscience postdoctoral scholar Barun K Saha, PhD student at IIT Kharagpur Jonathan Yates, IBM Watson Group EMEA Sam Richard, UI Architect, IBM Watson James Miller, Smith College, Author Singularity Rising Joel Pitt, Independent Researcher (ex-OpenCog) Achu Wilson, C.T.O Sastra Robotics Ji Tulach, CTO, Position s.r.o. Alexandru Litoiu, Yale University Mark Watson, Author and consultant specializing in artificial intelligence Michael Kuhlmann, Colony Networks George Kachergis, Postdoctoral researcher at New York University Brian Driscoll, Sr. Systems Engineer, Osprey Software Development Louis Choquel, Entrepreneur, Software Engineer Roberto Paura, Italian Institute for the Future Soheil Yasrebi, Loverino Inc. David Duvenaud, Harvard University James Babcock, Praxamed Peter Marshall, memememobile.com, CEO Marc Bejarano Igor Trajkovski, Time.mk Appu Shaji, Head, R&D, EyeEm Tim Daly, CMU LTI Senior Research Programmer Stefan Schubert, LSE Philosophy Colin Lewis, RobotEnomics Jared Peters, co-founder of Origamir Robotics Darryl McAdams, Language Engine Mike Slinn, Micronautics Research Tsvi Benson-Tilsen, University of Chicago, MIRI associate Nathaniel Thomas, Stanford University, PhD student in quantum computing Kyle Lussier, Founder / CEO of Tickle.me and Countervaillance Marek Rosa, CEO at Keen Software House Diana Hu, Data Scientist, OnCue TV Alejandro Machado, Carnegie Mellon University, graduate student Max Kesin, Palantir, ML developer Alexandros Marinos, CEO, Resin.io Patrick LaVictoire, MIRI research associate Michael Warner, AI researcher John Hering, Lookout Ronnie Vuine, Micropsi industries Chris Nicholson, Skymind Rene Verheij, AI programmer Rudy Krol, Amazon Web Services Simon Hughes, PhD Candidate Machine Learning, DePaul Aneesh Subramanian, University of Oxford Jon Baer, AI researcher James McDermott, University College Dublin Zavain Dar, VC and Lecturer Derek Brown, LinkedIn, Addepar Gabriel Synnaeve, Ecole Normale Suprieure / EHESS Denny Vrandecic, Google, Founder of Wikidata Robert W. Williams, Univ Tenn & Human Brain Project Peteris Erins, Consultant at McKinsey & Company Anubhav Ashok, University of Texas at Austin, Student and Apple Intern 2014 Naomi Moneypenny, AI Researcher & Chief Technology Officer, ManyWorlds, Inc David Cieslak, Aunalytics Stephan Zuchner, U of Miami, Professor and Chair for Human Genetics; Co-founder The Genesis Project and ViaGenetics Inc Evan Goldschmidt, Google Anna Salamon, Center for Applied Rationality Mark Koltko-Rivera, The Ontos Companies John Hammersley, co-founder of Overleaf / WriteLaTeX Malcolm Greaves, CMU Rob Bensinger, MIRI researcher Marcello Herreshoff, MIRI research associate, GooglePaul Pallaghy, Neo AI Systems P/L, Melbourne, Australia Percy Liang, Stanford, AI researcher Theresa Carbonneau, STG Gert de Cooman, Ghent University Nicholas Kong, Google Jeff Nelson, Founder, Chromebook project @ Google Christian Kaiser, Order of Magnitude Labs Gabriel Garrett, Artificial Consciousness Engineer Miles Brundage, Arizona State University Matthew Luciw, Boston University, Neurala, AI researcher Vijay Saraswat, IBM TJ Watson Research Center Ben Hamner, Chief Science Officer, Kaggle William Eden, Vice President, Thiel Capital Dan Von Kohorn, v2 Ratings Nicholas Haan, Singularity University Kristian Rnn, CEO and co-founder of Meta Mind AB, previously Projects Officer at the Future of Humanity Institute

To see the full list, click here.

Read the original:

AI Open Letter - Future of Life Institute

Samsung is preparing to launch a digital assistant called "Bixby," the latest product to result from the tech industry's obsession with artificial intelligence and the Internet of Things.

Bixby will be featured on the new Galaxy S8, Samsung's head of research and development Injong Rhee said in a blog post.

The S8 launches in New York next week.

Samsung is banking on the S8 to help it recover from last year's embarrassing Note 7 debacle. The company killed off the flagship device after a recall and various fixes failed to stop some Note 7s from overheating and catching fire.

It's also facing potential disruptions as de facto leader Lee Jaeyong's criminal trial begins in South Korea. Lee has been caught up in a corruption scandal and is facing a list of charges including bribery and embezzlement.

Bixby will enter a market that is already crowded with competitors, including Apple's (AAPL, Tech30) Siri, Amazon's (AMAZON) Alexa, Google (GOOG) Assistant, Microsoft's (MICROS) Cortana and IBM's (IBM, Tech30) Watson. Even Facebook (FB, Tech30) CEO Mark Zuckerberg has an assistant called Jarvis.

Samsung, however, insists that Bixby is "fundamentally different from other voice agents or assistants."

Related: What next for Samsung as chief's 'trial of the century' begins

The electronics giant said that Bixby's ability to work across supported apps sets it apart from Siri or Cortana. For example, you could direct BIxby to "find a photo of Jane and text it to Sally."

Users will also be able to switch between using Bixby to issue voice commands and using smartphones the old fashioned way, via touch commands. That is a clumsier experience on existing assistants, which often start tasks over if you switch from voice to touch.

Unlike its competitors, the S8 will come with a dedicated Bixby button, allowing users to fire up the smartphone digital assistant the same way they would a walkie-talkie. Samsung plans to make Bixby available on all its appliances, including air conditioners and TVs.

"We believe Bixby will evolve from a smartphone interface to an interface for your life," Rhee said.

Related: Roomba will now tell you what part of your home is dirtiest

Tech firms are betting that an increasing number of people will soon use digital assistants to interact with various devices. Research firm Tractica predicts the market for virtual digital assistants will top $15 billion by 2021.

Rhee said that Bixby would be "at the heart of our software and services evolution as a company."

Ian Fogg, a mobile devices analyst with IHS Markit, said the statement represented a major shift for the hardware giant.

"They've never made such a strong statement that they need to be a software and services company before," he said.

Some analysts remain skeptical of Bixby because it doesn't play to the firm's strengths. The company's other digital assistant, the S Voice, launched in 2012 and was quickly outpaced by Siri and Google Assistant.

"I am concerned about whether a traditionally hardware-centric company like Samsung can execute well on this, especially against ... heavyweights like Google," said Bryan Ma, a smartphone analyst with research firm IDC.

However, Ma said that even Apple hasn't perfected its digital assistant.

"It's still only the first inning of the ballgame right now," he said.

Last year, Samsung acquired a startup called Viv Labs in an effort to build its expertise in the area. Viv Labs is helmed by a co-creator of Apple's Siri, and its assistant can handle complex queries from users.

Bixby was reportedly developed using Samsung's in-house technology, but updates will incorporate Viv's features and tech.

CNNMoney (Hong Kong) First published March 21, 2017: 7:06 AM ET

Read the original post:

Samsung's new AI assistant will take on Siri and Alexa - CNNMoney

Artificial intelligence will make the electronic devices of tomorrow smarter, but not if their processorsmade by companies like ARMaren't up to the task.

Even though phones, smart TVs, and other connected devices aren't susceptible to the blue screen of death, they have countless other hardware and firmware limitations that chip-maker ARM is trying to solve.

The company announced a major overhaul of its chip microarchitecture this week, one that could boost the processing capabilities of everything from smart baby monitors to Fitbits to the next iPhone. Called Dynamiq, it is up to 50 times faster than ARM's existing architecture, which powers the current Cortex-A series of processors.

Why should you care about your phone's processing power? ARM executives point to the fact that phones and other devices of the future will be much smarter and more complex than today's crop of personal electronics, which means they'll need immense processing power to tackle all of their artificial intelligence algorithms. Even if the software is perfectly written and some of the number crunching is performed in the cloud, the device's own processor could still be a bottleneck.

"As systems get more complex, we need to redefine how multiprocessing works," ARM General Manager Nandan Nayampally said during a press briefing on Monday. "You will not be able to do this purely in the cloud."

And doing it on a device with today's processors will result in a problem that anyone who's tried a marathon virtual reality gaming session with their Samsung Gear VR has experienced: the phone will likely overheat and shut down. That equivalent of the blue screen of death might be little more than an inconvenience for gamers, but if it happened in a self-driving car, the consequences could be far more dire.

So Dynamiq is specifically designed to offer more performance while putting out less heat. It also supports AI and machine learning accelerators, a new class of microprocessor that can handle AI tasks while the main processor powers the phone's conventional tasks, such as taking photos or browsing the Internet. It's an evolution of ARM's "big.LITTLE" philosophy, which is all about choosing the right processor for the right task.

ARM says Dynamiq will also allow companies to certify their devices for the stringent ASIL-D standard that governs safety protocols for self-driving cars.

New chips based on the Dynamiq architecture will start showing up in consumer devices by 2018, Nayampally said. The company estimates that 100 billion ARM-based chips will be needed by 2021.

Tom is PCMag's San Francisco-based news reporter. He got his start in technology journalism by reviewing the latest hard drives, keyboards, and much more for PCMag's sister site, Computer Shopper. As a freelancer, he's written on topics as diverse as Borneo's rain forests, Middle Eastern airlines, and big data's role in presidential elections. A graduate of Middlebury College, Tom also has a master's journalism degree from New York University. Follow him on Twitter @branttom. More

Here is the original post:

ARM Announces Chip Overhaul for AI Future - PC Magazine

Forbes

From Automation To Empathy, AI Dominated The SXSW Conversation Forbes Beyond the political underpinning, the fake news agenda and the plethora of VR experiences, the one technology to really know about while at SXSW in Austin this year, was artificial intelligence (AI). That sentence has to be taken with a pinch of salt ...

Continued here:

From Automation To Empathy, AI Dominated The SXSW Conversation - Forbes

In early 2016, streaming giantNetflix, Inc. (NASDAQ: NFLX) announced that it had rolled out its service to 190 countries around the world. As the top provider of streaming content in the U.S., one of the biggest questions regarding the company's ability to succeed elsewhere was the issue of bandwidth. With slower internet speeds in many countries, would streaming performance suffer as a result?

With worldwide growth at stake, this was a question that the company needed to answer.

Turns out Netflix has a variety of tools it uses to navigate markets with underdeveloped bandwidth. Netflix CEO Reed Hastings was a keynote speaker at the 2017 Mobile World Congress in Barcelona, and in an interview with BBC broadcaster Francine Stock, he revealed some of the ways the company is addressing the issue.

Netflix AI tackles bandwidth issues. Image source: Pixabay.

The biggest revelation was regarding the use of artificial intelligence (AI). Netflix uses AI algorithms to review each frame of a video and compress it only to the degree necessary without degrading the image quality. This differs from previous technology that compressed the entire stream, but could cause fuzzy, pixelated or unclear images. This new method, which Netflix calls the Dynamic Optimizer, was developed to address bandwidth issues in emerging markets.

This not only improves streaming quality over slower speeds, but also tailors content for customers that view Netflix on tablets and phones, as is the case in countries like India, South Korea, and Japan. Providing video streaming at the same quality, but requiring lower bandwidth, also addresses the issue of data caps imposed locally by mobile providers.

Netflix collaborated with the University of Southern California and the University of Nantes in France to train the system, using hundreds of viewers and hundreds of thousands of scenes. By rating each scene individually on a variety of quality metrics, the AI system learned to determine image quality. Hastings described the technological advancement like this:

What we've done is invest in the codex, the video encoders, so that at a half a megabit, you get incredible picture quality on a 4- and 5-inch screen. Now, we're down in some cases to 300 kilobits and we're hoping someday to be able to get to 200 kilobits for an amazing picture. So we're getting more and more efficient at using operators' bandwidth.

Innovative solutions to technical problems. Image source: Netflix, Inc.

Hastings explained that the company was investing in many other ways to make buffering a thing of the past. He stated that the company was working on interconnect agreements with internet service providers (ISPs) across the globe, which provide increased speeds as the result of a more direct connection. Netflix has developed its own content delivery network in a program it calls Open Connect. It provides specialized Netflix servers directly to the ISPs, whose sole function is to deliver Netflix content to local subscribers.

Netflix has been working on tailored solutions for video encoding for a number of years. In a Netflix Tech Blogentry from 2015, the company described a complex method for analyzing each title and assigning it a specific encoding rate based on the genre and complexity of the scenery. An action movie might have significant motion, fast-moving objects, rapid scene changes, explosions, and water splashes versus an animated title that produces significantly less distortion at the same bandwidth. These variables were considered across the company's vast library of TV shows and movies and each title was assigned its own compression rate, thereby providing the best quality at the lowest bandwidth.

As competition increases and Netflix continues its expansion into less-developed markets, technological innovations such as these will provide a key competitive advantage. Rather than relying on local market forces and providers to dictate terms, the company is using novel solutions to address its challenges.

10 stocks we like better than Netflix

When investing geniuses David and Tom Gardner have a stock tip, it can pay to listen. After all, the newsletter they have run for over a decade, Motley Fool Stock Advisor, has tripled the market.*

David and Tom just revealed what they believe are the 10 best stocks for investors to buy right now... and Netflix wasn't one of them! That's right -- they think these 10 stocks are even better buys.

*Stock Advisor returns as of February 6, 2017

Danny Vena owns shares of Netflix and watches for too much content. The Motley Fool owns shares of and recommends Netflix. The Motley Fool has a disclosure policy.

Read more from the original source:

Netflix Is Using AI to Conquer the World... and Bandwidth Issues - Bloomington Pantagraph

Forbes

How AI Can Prove Workers' Best Defense In The Race Against Automation Forbes United Technology's announcement last November that its Carrier Corp. plant would keep jobs in Indiana rather than move them to Mexico was heralded as a significant victory for American workers. However, the true impact of the deal was hidden below the ...

Read more:

How AI Can Prove Workers' Best Defense In The Race Against Automation - Forbes

From voice translation to self-driving automobile, AI's impact in everyday life will become more and more apparent this year. The AI and deep learning market will experience even more rapid technological advancement, very rapid growth and adoption, and increasing competition for both hardware and software platforms. While AI fears will remain, the public will become more cognisant and comfortable with social media AI applications.

Deep learning training lends itself to what we call "High Density Processing". High density processing applies when algorithms are computationally intensive, having higher ratios of compute operations per byte of memory bandwidth.

In such cases dense clusters of multicore CPUs hosting accelerator technology can provide highly favourable cost-performance and performance per watt. GPUs, because of their ability to provide high density processing, have enabled deep learning computations and have dominated recently.

In 2017, we will start seeing a move from the near monopoly of GPUs for training to hosting on a rather wide variety of multicore and accelerator technologies. These will include Knights Landing/Knights Mill chips and AI accelerators implemented as FPGAs or ASICs. But the GPU will still be widely used.

Short fixed-point arithmetic can offer order of magnitude performance advantages over floating point (see here). These low-power solutions with special purpose architectures can demonstrate better price/performance than even GPUs.

Cloudy options abound with Amazon, Baidu and Microsoft having augmented their GPU-based cloud offerings with FPGA options for AI applications, and Google "supercharging" their Cloud AI with ASICs known as Tensor Processing Units employing short arithmetic.

Intel will also be a leader in bringing accelerators to market, and the combination of Knights Landing plus the Nervana Engine technology to be unveiled later this year looks particularly intriguing.

So in 2017, GPU dominance will be eroded. We think it is premature to talk about a "post-GPU" era, and expect GPUs to maintain a very comfortable lead, but we do expect a much richer mix of technologies to emerge.

Deep learning isn't just about the hardware; software libraries that enable algorithms that take advantage of said hardware and that put the technology into more hands are maybe even more important. We're seeing several libraries battle for dominance in the AI arena. Google's TensorFlow has leapt to the forefront on GitHub and Intel recently responded with their BigDL deep learning framework for Spark. Theano, Microsoft's CNTK and many others the vast majority of which have CUDA support will compete eagerly for developer mindshare. It's too early to call the race, but our prediction is that Microsoft and Intel are the most likely companies to give Google a run for the money.

What's in it for Microsoft is promotion of their software ecosystem, especially around Big Data and IoT. Intel wants increased hardware sales, not surprisingly. And Google appears to be most interested in growing their developer ecosystem to gather new applications that they can then monetise in areas such as self-driving automobiles.

Voice translation will be one of the biggest breakout application segments. International travellers will begin to use it regularly on their mobile phones for short conversations, including ordering food and coffee, buying train tickets, and other shopping.

Text translation in messaging apps has become routine, especially for certain language pairs, facilitating communication between lovers, family and friends, and international project team members.

The many "Lost In Translation" occurrences lead to abundant laughter, frustration and misunderstandings, and even breakups. Despite the limitations of machine translation, the appeal will be irresistible. Usage in personal social interaction will initially be much greater than for business. Could this be the killer app for consumer AI?

AI fears in some respects will ease as the public becomes more cognisant and comfortable with AI applications that are accessed from, or that support applications running on, their mobile-based social media platforms (Google, Facebook, and Twitter in particular).

But concerns around governments' electronic monitoring of social media content and face recognition in public spaces will remain. Facebook, Twitter and others will struggle with the appropriate level of tuning of AI solutions to filter out fake news, offensive videos, and hate speech. Their complicity with nondemocratic government requests for censorship will grow at the expense of freedom of expression. In addition, concerns around middle-class job losses to automated machine learning systems will continue to grow as the globalisation backlash continues.

A raft of AI applications in healthcare including for diagnosis, patient monitoring, and even clinical trials will make steady progress, but there will be no major breakthroughs.

Patient interest will grow significantly as AI healthcare case studies become more numerous and positive outcomes recorded. AI will be seen in a very positive light for medical imaging evaluation and diagnosis, and will begin to lead to significant cost savings. Trial usage for laboratory tests (blood, urine) will grow, but lag usage for imaging applications significantly.

Although patients might be concerned about robots replacing doctors, it will be enhancement, not replacement that is relevant. Generally the patient will not know when doctors and nurses are using AI to support healthcare decisions.

As one example, IBM's Watson technology reached the same diagnosis as oncologists in 99 per cent of cancer cases examined, yet it was also able to explore a wider range of options, since it can extremely rapidly explore the medical literature. Second opinions will be arrived at in realtime, which will save everyone time and money.

Self-driving automobile ("auto-automobile") technology will advance, but will experience speed bumps and citizen backlash as a growing number of trials leads to more accidents, including fatalities, even as statistics point to a significant reduction in accidents. Local and national government restrictions will tighten, and trials will increasingly be focused on lower-risk driving scenarios.

In the high-risk arena, military interest in self-driving ground-based vehicles will become very evident, due to the potential savings of lives and money, and the prospects for using such vehicles to confuse the enemy since soldier casualties will be removed from the equation.

So, to sum up: You're not going to exclusively use GPUs as your AI engine forever, and you're going to have a wide range of choices when it comes to AI libraries. You'll be using AI language translators for pick-up lines on your next international business trip, making you more comfortable with AI applications, but you'll still be afraid of what the government might do with the same technology, and that an AI might take your job.

You'll be healthier because AI medical care applications will speedily diagnose and recommend treatments for your injuries and ills. And, finally, there's a slightly better chance you'll need this enhanced medical care since self-driving cars will be tested in greater numbers. Phew.

See the original post:

What 2017 holds for AI: Will you fear or embrace our machine overlords? - The Register

Jessica Pothering

Jessica is a business and finance writer, focusing on impact investing, social entrepreneurship and economic development. She previously reported for financial publications covering the global private equity, real estate and insurance markets.

AI see, AI do.

Among the risks in using data-driven AI in low-resource or at-risk communities is that the algorithms will magnify systemic biases.

Care must be taken to prevent AI systems from reproducing discriminatory behavior, such as machine learning that identifies people through illegal racial indicators, write researchers from the Stanford One Hundred Study on Artificial Intelligence.

This week, ImpactAlpha is extracting nuggets from Stanfords century-long effort to understand AIs long-term possibilities and dangers. Theres already an update to yesterdays #2030 segment on self-driving cars: Ford recently announced a $1 billion investment in software for autonomous fleets.

The researchers found AI could be a money-saving lifeline for budget-strapped local and state governments.

Illinois Department of Human Services, for example, is using predictive data modelling to improve prenatal care to high-risk pregnant women.

Cincinnati is using AI to identify and inspect properties that arent up to code.

AI also has potential for developing low-cost community health campaigns, which are otherwise difficult to target and expensive to implement.

This post originally appeared in ImpactAlphas daily newsletter.Get TheBrief.

Photo credit: Scienceofsingularity.com

See original here:

Underserved communities: Leveraging AI to improve health and social services in low-resource areas - ImpactAlpha (registration)

Atul Kumar, Chief Product Officer at Mintigo suggests marketers what they can achieve realistically using AI in 2017

2016 was a tremendous year for MarTech. According to Forrester Tech Radar for B2B Marketing Technology and SD16 (SiriusDecisions conference 2016), two of the hottest trends last year were ABM (Account-Based Marketing) and Predictive Marketing. Im sure you all are now busy with deploying your ABM initiatives. Many of you have embraced or thinking of Predictive Marketing.

AI is the red hot topic that is being discussed at c-level in all organizations. We all are beneficiaries of AI in our daily lives; from Alexa and Siri to TacoBot, we are reaping the rewards of AI. Googles AI (AlphaGo) beat the world master in the game GO. In the later half of 2016, Salesforce.com announced Salesforce Einstein while Oracle promised to do better with applications built with Adaptive Intelligence. And self-driving cars are coming soon! Its clear that AI powered business and consumer applications have arrived and you need to be ready to have an intelligent conversation with your boss and peers.

Im sure youve heard of many different terms such as AI, predictive analytics, machine learning, neural networks, deep learning etc. all used interchangeably by the industry. AI is an umbrella term, a branch of computer science whereas machine learning, deep learning etc. are some of the methods and systems of enabling AI. For example, the virtual assistants or Bots, excellent examples of AI, actually use NLP/G (natural language processing/generation) to understand and respond to human requests. There is no need to panic when you hear different terms or some vendor try to tell you that we do AI and others dont! Whats more important that you understand what AI can do for you.

Here is what you can achieve realistically using AI in 2017:

1. Account and lead selection - AI can help you select your best accounts and leads for your inbound, outbound and ABM initiatives. AI platforms such as the one my company (Mintigo) offers, help you to build predictive models for any business scenario, from cross/up-sell to new product launches. You can then explore your total available market using discovery tools offered by these platforms. You can also get new lead names (look-alikes) for your campaigns as needed based on predictive insights.

2. Personalized 1:1 Nurtures - One of the key outputs of predictive analytics is a set of attributes that represents the model. These attributes, often referred to as ideal customer DNA or profile, defines the why or the reasoning of a predictive score. Why John Smith @IBM is more likely to buy your product or services as compared to Andy Cheng @HP. By comparing these sets of attributes for the two, you can easily understand the reasoning. Some of the AI platforms have the ability to provide you detailed attributes into your MAP & CRM systems in real-time. This allows you to nurture an inbound inquiry or send an outbound message in a very personalized manner; for example, if your ideal customer DNA includes modern marketers who spend more than average on digital marketing and use one of the marketing automation systems, you can engage your prospects by sending them offers and messages that are relevant to their needs. AI eliminates the generic messages, and engage your prospects with the right (and relevant) message at the right time.

3. Automated Campaigns - This is an advanced application where an AI application automatically builds the customer journey to accelerate time to sales. This is accomplished by engaging the right prospects with the right offers using the right channels at the right time. Simply put, it is the martech nirvana that only AI can enable. A good percentage of MAP users still use a single step email nurture (nice way to say batch and blast!). Lack of adoption of multi-step nurture can be attributed to lack of resources and time it takes to create one. Those who do run multi-step nurtures, are doing so with limited data and utilizing the art of marketing. The Automated AI driven campaigns solve this issue. However, to take advantage of this you need to organize your content and track each LP uniquely (channel, offer, offer type) using utm codes or equivalent. The target and response data must be available or use a system that offers test and learn abilities.

4. Sales Engagement - Helping sales to intelligently engage with the precious leads and accounts is crucial for the success of your business. AI applications are changing the conventional B2B sales in many different ways; automated conversational Bots (such as Conversica) and other task management bots are good early examples. Other AI applications, such as the one from Mintigo (Mintigos Predictive Sales Coach), are enabling sales to engage their prospects intelligently. These applications use artificial intelligence (AI) to identify the Who, What, Why, and How? of the sales process. To be more specific, Who will buy from your company, What will they buy from you, Why they need it, and How you should engage them in a meaningful conversation. Selling is a tough business and require tremendous amount of research despite enormous efforts put in enabling sales. Many enterprises hire fresh grads to dial for the dollars; the churn could be high if they were not continuously enabled. Marketing creates content that is often buried behind corporate firewalls in content stores. What is needed is always-on intelligence and messaging/content that enables sales to easily find the right prospect and then quickly understand his/her needs and has messages and content at their fingertips to have a meaningful conversation. AI applications are here to help you not only build trust relationship with your counterparts in sales but impact revenue directly.

This list is by no means complete or even close. Numerous AI driven applications are being born everyday. This provides a starting point that will help you navigate the world of AI for marketing. And finally, dont forget to watch ex machina to explore the art of possibility.

See the original post:

AI for B2B Marketers What to Expect in 2017? - MarTech Advisor