Menu

▼

Saturday, January 22, 2022

g-f(2)846 THE NEW WORLD NEWS (1/22/2022), MIT Technology Review, Meta’s new learning algorithm can teach AI to multi-task


VIRAL KNOWLEDGE: The “genioux facts” knowledge news


The New World NEWS

If you only have some seconds




ULTRA-condensed knowledge


"g-f" fishing of golden knowledge (GK) of the fabulous treasure of the digital age, The New World News, General-purpose AI (1/22/2022)  g-f(2)426 


Opportunity, MIT Technology Review

EXCEPTIONAL “Full Pack Golden Knowledge Container”


Meta’s new learning algorithm can teach AI to multi-task


  • The single technique for teaching neural networks multiple skills is a step towards general-purpose AI.
  • Deep neural networks have become very good at identifying objects in photos and conversing in natural language, but not at the same time: there are AI models that excel at one or the other, but not both. 
  • Part of the problem is that these models learn different skills using different techniques. 
  • A team at Meta AI (previously Facebook AI Research) wants to change that. 


      Genioux knowledge fact condensed as an image




      ABOUT THE AUTHORS


      Will Douglas Heaven


      Senior editor, AI

      I (@strwbilly, LinkedIn) am the senior editor for AI at MIT Technology Review, where I cover new research, emerging trends and the people behind them. Previously, I was founding editor at the BBC tech-meets-geopolitics website Future Now and chief technology editor at New Scientist magazine. I have a PhD in computer science from Imperial College London and know what it’s like to work with robots.


      Extra-condensed knowledge


      Lessons learned, MIT Technology Review

      Golden Knowledge (GK) juice


      • The researchers have developed a single algorithm that can be used to train a neural network to recognize images, text, or speech. The algorithm, called Data2vec, not only unifies the learning process but performs at least as well as existing techniques in all three skills. “We hope it will change the way people think about doing this type of work,” says Michael Auli, a researcher at Meta AI.
      • Data2vec is part of a big trend in AI toward models that can learn to understand the world in more than one way. “It’s a clever idea,” says Ani Kembhavi at the Allen Institute for AI in Seattle, who works on vision and language. “It’s a promising advance when it comes to generalized systems for learning.”
      • The researchers were surprised to find that their approach actually performed better than existing techniques at recognizing images and speech, and performed as well as leading language models on text understanding.


      Condensed knowledge




      Learned lessons, Meta AI


      The first high-performance self-supervised algorithm that works for speech, vision, and text

      Golden Knowledge (GK) juice


      • Self-supervised learning — where machines learn by directly observing the environment rather than being explicitly taught through labeled images, text, audio, and other data sources — has powered many significant recent advances in AI. But while people appear to learn in a similar way regardless of how they get information — whether they use sight or sound, for example — there are currently big differences in the way self-supervised learning algorithms learn from images, speech, text, and other modalities.
      • This discrepancy has been a significant barrier to applying advances in self-supervised learning more broadly. Because a powerful algorithm designed for, say, understanding images can’t be directly applied to another modality, such as text, it is difficult to push several modalities ahead at the same rate.
      • This is why Meta AI developed and is excited to announce data2vec, the first high-performance self-supervised algorithm that works for multiple modalities. We apply data2vec separately to speech, images and text and it outperformed the previous best single-purpose algorithms for computer vision and speech and it is competitive on NLP tasks. It also represents a new paradigm of holistic self-supervised learning, where new research improves multiple modalities rather than just one. It also does not rely on contrastive learning or reconstructing the input example. In addition to helping accelerate progress in AI, data2vec brings us closer to building machines that learn seamlessly about different aspects of the world around them. It will enable us to develop more adaptable AI, which we believe will be able to perform tasks beyond what today’s systems can do.
      • As part of this announcement, we are sharing code and pretrained models on data2vec so that others in the research community can build upon our work.
      • How data2vec works
      • Toward machines that learn from observing the world around them

      References



      This blog post was made possible by the work of Alexei Baevski, Wei-Ning Hsu, Qiantong Xu, Arun Babu, Jiatao Gu, Michael Auli.



      Some relevant characteristics of this "genioux fact"

      • Category 2: The Big Picture of the Digital Age
      • [genioux fact deduced or extracted from MIT Technology Review + Meta AI]
      • This is a “genioux fact fast solution.”
      • Tag Opportunities those travelling at high speed on GKPath
      • Type of essential knowledge of this “genioux fact”: Essential Analyzed Knowledge (EAK).
      • Type of validity of the "genioux fact". 

        • Inherited from sources + Supported by the knowledge of one or more experts.


      References


      “genioux facts”: The online programme on MASTERING “THE BIG PICTURE OF THE DIGITAL AGE”, g-f(2)846, Fernando Machuca, January 22, 2022, blog.geniouxfacts.com, geniouxfacts.com, Genioux.com Corporation.


      ABOUT THE AUTHORS


      PhD with awarded honors in computer science in France

      Fernando is the director of "genioux facts". He is the entrepreneur, researcher and professor who has a disruptive proposal in The Digital Age to improve the world and reduce poverty + ignorance + violence. A critical piece of the solution puzzle is "genioux facts". The Innovation Value of "genioux facts" is exceptional for individuals, companies and any kind of organization.



      Key “genioux facts”


      • Some relevant and recent "genioux facts" from the g-f golden knowledge pyramid to master the big picture of the digital age. 
        1. g-f(2)845 THE NEW WORLD NEWS (1/22/2022), Bloomberg, Bezos Loses $20 Billion in Brutal Week for Tech Billionaires
        2. g-f(2)844 THE NEW WORLD NEWS (1/22/2022), WSJ, The Nanotechnology Revolution Is Here—We Just Haven’t Noticed Yet
        3. g-f(2)843 THE BIG PICTURE OF THE DIGITAL AGE (1/22/2022), The New York Times, An article as a brilliant, profound and meaningful work of art
        4. g-f(2)842 THE BIG PICTURE OF THE DIGITAL AGE (1/21/2022), HBR, Leadership Training Shouldn’t Just Be for Top Performers
        5. g-f(2)841 THE BIG PICTURE OF THE DIGITAL AGE (1/21/2022), MIT SMR, Why Companies Practice Corporate Social Responsibility
        6. g-f(2)840 The New World (1/21/2022), Forbes, Artificial Intelligence: A Key Technology That's Shaping Our Tomorrow
        7. g-f(2)839 THE NEW WORLD (1/21/2022), WSJ, Explore the Top 250 Best-Managed Companies of 2021
        8. g-f(2)838 THE BIG PICTURE OF YOUR LIFE (1/20/2022), geniouxfacts, The Playbook for a Successful Personal Digital Transformation (PDT)
        9. g-f(2)837 Lighthouse of the big picture of the digital age (1/20/2022), Essential FACT 5: Becoming a fast and successful learner is a superpower
        10. g-f(2)836 THE BIG PICTURE OF THE DIGITAL AGE (1/19/2022), Media, More reactions to Microsoft's strategic move
        11. g-f(2)835 THE BIG PICTURE OF THE DIGITAL AGE (1/19/2022), The New York Times, Millions Have Lost a Step Into the Middle Class, Researchers Say
        12. g-f(2)834 THE NEW WORLD (1/19/2022), WSJ + Media, Microsoft’s Activision Blizzard Deal to Power Its Netflix-of-Gaming Aspirations
        13. g-f(2)833 THE BIG PICTURE OF THE DIGITAL AGE (1/19/2022), INSEAD Knowledge, What Distinguishes the Super Rich From the Rest of Us
        14. g-f(2)832 THE BIG PICTURE OF THE DIGITAL AGE (1/19/2022), HBR, How Digital Transformation Can Improve Hospitals’ Operational Decisions
        15. g-f(2)831 THE BIG PICTURE OF THE DIGITAL AGE (1/18/2022), Talks at Google, Dorie Clark | How to Be a Long Term Thinker in a Short Term World
        16. g-f(2)830 The New World (1/18/2022), TechXplore, Controlling complex systems with Artificial Intelligence
        17. g-f(2)829 The New World (1/18/2022), Forbes, 3 Ways That Artificial Intelligence (AI) Will Change Your Job Forever
        18. g-f(2)828 THE BIG PICTURE OF THE DIGITAL AGE (1/18/2022), HBR, When Lower Intensity Leads to Higher Results
        19. g-f(2)827 THE BIG PICTURE OF THE DIGITAL AGE (1/17/2022), HBR, Research: How Entrepreneurship Can Revitalize Local Communities
        20. g-f(2)826 The New World (1/17/2022), Bloomberg, France Gets its 25th Tech Unicorn, Reaching Macron’s Goal Early
        21. g-f(2)825 The New World (1/17/2022), Forbes, 14 Ways To Leverage AI To Better Connect With Customers
        22. g-f(2)824 The New World (1/17/2022), Venture Beat, How AI will drive the hybrid work environment
        23. g-f(2)823 The New World (1/17/2022), Forbes, How To Close More Deals With Artificial Intelligence
        24. g-f(2)822 THE BIG PICTURE OF THE DIGITAL AGE (1/16/2022), geniouxfacts, The 10 most read "genioux facts" of the month
        25. g-f(2)821 THE BIG PICTURE OF THE DIGITAL AGE (1/16/2022), geniouxfacts, The 10 most read "genioux facts" of the week
        26. g-f(2)820 The New World (1/15/2022), Venture Beat, How AI and language intelligence are helping employees learn
        27. g-f(2)819 The New World (1/15/2022), MIT SMR, Toxic Culture Is Driving the Great Resignation
        28. g-f(2)818 The New World (1/14/2022), Thomas M. Siebel, Digital Transformation: Survive and Thrive in an Era of Mass Extinction
        29. g-f(2)817 The New World (1/14/2022), HBR Analytical Services, Keys to Successful Innovation through Artificial Intelligence
        30. g-f(2)816 The New World (1/14/2022), HBR Analytic Services, AWS, AWS perspective: AI have the potential to transform nearly every industry
        31. g-f(2)815 The New World (1/14/2022), WEF, Artificial intelligence: What the C-suite needs to know
        32. g-f(2)814 The New World (1/14/2022), Analytics Insight, 10 WAYS ARTIFICIAL INTELLIGENCE WILL CHANGE THE WORLD IN 2022
        33. g-f(2)813 The New World News (1/13/2022), WSJ, These TikTok Stars Made More Money Than Many of America’s Top CEOs
        34. g-f(2)812 The New World (1/13/2022), Brookings, How countries are leveraging computing power to achieve their national artificial intelligence strategies
        35. g-f(2)811 The New World (1/13/2022), Bloomberg, TSMC to Spend at Least $40 Billion to Address Chip Shortages
        36. g-f(2)810 THE BIG PICTURE OF THE DIGITAL AGE (1/12/2022), Talks at Google, Bruce Hood | How Minds are Constructed
        37. g-f(2)809 THE NEW WORLD (1/12/2022), geniouxfacts, 10 juices of ultra-condensed GK to keep pace
        38. g-f(2)808 THE BIG PICTURE OF THE DIGITAL AGE (1/12/2022), MIT SMR + BCG, AI in the Supply Chain: Cold Chain Technologies’ Ranjeet Banerjee
        39. g-f(2)807 THE NEW WORLD (1/11/2022), HBR, Finding Success Starts with Finding Your Purpose
        40. g-f(2)806 The New World (1/11/2022), HBR, HBR’s Most-Read Research Articles of 2021
        41. g-f(2)805 The New World (1/11/2022), msn money, The National, Artificial intelligence to influence top tech trends in major way in next five years
        42. g-f(2)804 The New World (1/11/2022), MIT SMR, Six Ways Leaders Can Adapt to the Workplace of 2022