How to Learn an Algorithm - Jürgen Schmidhuber

Описание: Video of a talk by Jürgen Schmidhuber at our third Deep Learning in Action talk series 2015-10-07.
http://munichacm.de/deeplearning/

The event's description can be found
at http://www.meetup.com/en/deeplearning...

Abstract: "Since 1987 I have published research on the “one algorithm” or “master program” that explains intelligence, predicting that in hindsight it will seem so simple that high school students will be able understand and implement it. Here I review 30 years of our work on both gradient-based and more general problem solvers that search the space of algorithms running on general purpose computers with internal memory. Architectures include traditional computers, Turing machines, recurrent neural networks, fast weight networks, and stack machines. Some of our algorithm searchers are based on algorithmic information theory and are asymptotically optimal. Most can learn to direct internal and external spotlights of attention. Some are self-referential and can even learn the learning algorithm itself (recursive self-improvement). Some can solve very deep algorithmic problems (involving billions of steps) that are infeasible using more recent memory-based deep learners. Recently, algorithms learned by our Long Short-Term Memory recurrent networks have defined the state-of-the-art in handwriting recognition, speech recognition, natural language processing, machine translation, and image caption generation. Google and other companies have made them available to over a billion users."

Most of the slides (including slides for other recent talks): http://people.idsia.ch/~juergen/deep2...


Bio: Since age 15 or so, Prof. Jürgen Schmidhuber's main goal has been to build a self-improving Artificial Intelligence (AI) smarter than himself, then retire. He has pioneered self-improving general problem solvers since 1987, and Deep Learning Neural Networks (NNs) since 1991. The recurrent NNs (RNNs) developed by his research groups at the Swiss AI Lab IDSIA & USI & SUPSI and TU Munich were the first RNNs to win official international contests. They have revolutionised connected handwriting recognition, speech recognition, machine translation, optical character recognition, image caption generation, and are now used by Google, Microsoft, IBM, Baidu, and many other companies. Founders & staff of DeepMind (sold to Google for over 600M) include 4 former PhD students from his lab. His team's Deep Learners were the first to win object detection and image segmentation contests, and achieved the world's first superhuman visual classification results, winning nine international competitions in machine learning & pattern recognition (more than any other team). They also were the first to learn control policies directly from high-dimensional sensory input using reinforcement learning. His research group also established the field of mathematically rigorous universal AI and optimal universal problem solvers. His formal theory of creativity & curiosity & fun explains art, science, music, and humor. He also generalized algorithmic information theory and the many-worlds theory of physics, and introduced the concept of Low-Complexity Art, the information age's extreme form of minimal art. Since 2009 he has been member of the European Academy of Sciences and Arts. He has published 333 peer-reviewed papers, earned seven best paper/best video awards, the 2013 Helmholtz Award of the International Neural Networks Society, and the 2016 IEEE Neural Networks Pioneer Award. He is president of NNAISENSE, which aims at building the first practical general purpose AI.

His website can be found at http://people.idsia.ch/~juergen/.