Jürgen Schmidhuber

Jürgen Schmidhuber
Born (1963-01-17) January 17, 1963
Munich, Germany
Residence Switzerland
Nationality German
Fields Artificial intelligence
Institutions Dalle Molle Institute for Artificial Intelligence Research
Alma mater Technische Universität München
Thesis Dynamische neuronale Netze und das fundamentale raumzeitliche Lernproblem
(Dynamic neural nets and the fundamental spatio-temporal credit assignment problem) (1990)

Jürgen Schmidhuber (born 17 January 1963 in Munich) is a German computer scientist and artist known for his work on machine learning, Artificial Intelligence (AI), artificial neural networks, digital physics, and low-complexity art. His contributions also include generalizations of Kolmogorov complexity and the Speed Prior. From 2004 to 2009 he was professor of Cognitive Robotics at the Technische Universität München. Since 1995 he has been co-director of the Swiss AI Lab IDSIA in Lugano, since 2009 also professor of Artificial Intelligence at the University of Lugano. Between 2009 and 2012, the recurrent neural networks and deep feedforward neural networks developed in his research group have won eight international competitions in pattern recognition and machine learning.[1] More than a billion people can now use his lab's algorithms, for example, through Google's speech recognition for smartphones,[2] which is now based on a deep learning method called Long short term memory developed since 1995. In honor of his achievements he was elected to the European Academy of Sciences and Arts in 2008. He received numerous awards including the 2013 Helmholtz Award of the International Neural Networks Society. He also is recipient of the 2016 IEEE CIS Neural Networks Pioneer Award, with a citation for "pioneering contributions to deep learning and neural networks."[3]

Contributions

Recurrent neural networks

The dynamic recurrent neural networks developed in his lab are simplified mathematical models of the biological neural networks found in human brains. A particularly successful model of this type is called Long short term memory.[4] From training sequences it learns to solve numerous tasks unsolvable by previous such models. Applications range from automatic music composition to speech recognition, reinforcement learning and robotics in partially observable environments. As of 2010, his group has the best results on benchmarks in automatic handwriting recognition, obtained with deep neural networks[5] and recurrent neural networks.[6] Hundreds of millions of users can now use LSTM for numerous applications, including Google's speech recognition on smartphones.[2]

Artificial evolution / genetic programming

As an undergrad at TUM Schmidhuber evolved computer programs through genetic algorithms. The method was published in 1987 as one of the first papers in the emerging field that later became known as genetic programming. In the same year he published the first work on Meta-genetic programming. Since then he has co-authored numerous additional papers on artificial evolution. Applications include robot control, soccer learning, drag minimization, and time series prediction. He received several best paper awards at scientific conferences on evolutionary computation.

Neural economy

In 1989 he created the first learning algorithm for neural networks based on principles of the market economy (inspired by John Holland's bucket brigade algorithm for classifier systems): adaptive neurons compete for being active in response to certain input patterns; those that are active when there is external reward get stronger synapses, but active neurons have to pay those that activated them, by transferring parts of their synapse strengths, thus rewarding "hidden" neurons setting the stage for later success.[7]

Artificial curiosity and creativity

In 1990 he published the first in a long series of papers on artificial curiosity and creativity for an autonomous agent. The agent is equipped with an adaptive predictor trying to predict future events from the history of previous events and actions. A reward-maximizing, reinforcement learning, adaptive controller is steering the agent and gets curiosity reward for executing action sequences that improve the predictor. This discourages it from executing actions leading to boring outcomes that are either predictable or totally unpredictable.[8] Instead the controller is motivated to learn actions that help the predictor to learn new, previously unknown regularities in its environment, thus improving its model of the world, which in turn can greatly help to solve externally given tasks. This has become an important concept of developmental robotics. Schmidhuber argues that his corresponding formal theory of creativity explains essential aspects of art, science, music, and humor.[9]

Unsupervised learning / factorial codes

During the early 1990s Schmidhuber also invented a neural method for nonlinear independent component analysis (ICA) called predictability minimization. It is based on co-evolution of adaptive predictors and initially random, adaptive feature detectors processing input patterns from the environment. For each detector there is a predictor trying to predict its current value from the values of neighboring detectors, while each detector is simultaneously trying to become as unpredictable as possible.[10] It can be shown that the best the detectors can do is to create a factorial code of the environment, that is, a code that conveys all the information about the inputs such that the code components are statistically independent, which is desirable for many pattern recognition applications.

Kolmogorov complexity / computer-generated universe

In 1997 Schmidhuber published a paper based on Konrad Zuse's assumption (1967) that the history of the universe is computable. He pointed out that the simplest explanation of the universe would be a very simple Turing machine programmed to systematically execute all possible programs computing all possible histories for all types of computable physical laws.[11][12] He also pointed out that there is an optimally efficient way of computing all computable universes based on Leonid Levin's universal search algorithm (1973). In 2000 he expanded this work by combining Ray Solomonoff's theory of inductive inference with the assumption that quickly computable universes are more likely than others.[13] This work on digital physics also led to limit-computable generalizations of algorithmic information or Kolmogorov complexity and the concept of Super Omegas, which are limit-computable numbers that are even more random (in a certain sense) than Gregory Chaitin's number of wisdom Omega.[14]

Universal AI

Important research topics of his group include universal learning algorithms and universal AI[15][16] (see Gödel machine). Contributions include the first theoretically optimal decision makers living in environments obeying arbitrary unknown but computable probabilistic laws, and mathematically sound general problem solvers such as the remarkable asymptotically fastest algorithm for all well-defined problems, by his former postdoc Marcus Hutter. Based on the theoretical results obtained in the early 2000s, Schmidhuber is actively promoting the view that in the new millennium the field of general AI has matured and become a real formal science.

Low-complexity art / theory of beauty

Schmidhuber's low-complexity artworks (since 1997) can be described by very short computer programs containing very few bits of information, and reflect his formal theory of beauty[17] based on the concepts of Kolmogorov complexity and minimum description length.

Schmidhuber writes that since age 15 or so his main scientific ambition has been to build an optimal scientist, then retire. First he wants to build a scientist better than himself (he quips that his colleagues claim that should be easy) who will then do the remaining work. He claims he "cannot see any more efficient way of using and multiplying the little creativity he's got".

Robot learning

In recent years a robotics group with focus on intelligent and learning robots, especially in the fields of swarm and humanoid robotics was established at his lab.[18] The lab is equipped with a variety of mobile and flying robots and is one of the around 20 labs in the world owning an iCub humanoid robot. The group has applied a variety of machine learning algorithms, such as reinforcement learning and genetic programming, to improve adaptiveness and autonomy of robotic systems.

Recently his work on evolutionary robotics, with a focus on using genetic programming to evolve robotic skills, especially in robot vision have allowed for quick and robust object detection in humanoid robots.[19][20][21] IDSIA's work with the iCub humanoid won the 2013 AAAI Student Video competition.[22][23]

References

  1. ↑ 2012 Kurzweil AI Interview with Jürgen Schmidhuber on the eight competitions won by his Deep Learning team 2009-2012
  2. 1 2 H. Sak, A. Senior, K. Rao, F. Beaufays, J. Schalkwyk (2015): Google voice search: faster and more accurate. http://googleresearch.blogspot.ch/2015/09/google-voice-search-faster-and-more.html
  3. ↑ IEEE Computational Intelligence Society - Award Recipients: http://cis.ieee.org/award-recipients.html
  4. ↑ S. Hochreiter and J. Schmidhuber. Long Short-Term Memory. Neural Computation, 9(8):1735–1780, 1997.
  5. ↑ D. C. Ciresan, U. Meier, L. M. Gambardella, J. Schmidhuber. Deep Big Simple Neural Nets For Handwritten Digit Recognition. Neural Computation 22(12): 3207-3220.
  6. ↑ A. Graves, M. Liwicki, S. Fernandez, R. Bertolami, H. Bunke, J. Schmidhuber. A Novel Connectionist System for Improved Unconstrained Handwriting Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 31, no. 5, 2009.
  7. ↑ J. Schmidhuber. A local learning algorithm for dynamic feedforward and recurrent networks. Connection Science, 1(4):403–412, 1989
  8. ↑ J. Schmidhuber. Curious model-building control systems. In Proc. International Joint Conference on Neural Networks, Singapore, volume 2, pages 1458–1463. IEEE, 1991
  9. ↑ J. Schmidhuber. Formal Theory of Creativity, Fun, and Intrinsic Motivation (1990–2010). IEEE Transactions on Autonomous Mental Development, 2(3):230–247, 2010.
  10. ↑ J. Schmidhuber. Learning factorial codes by predictability minimization. Neural Computation, 4(6):863–879, 1992
  11. ↑ J. Schmidhuber. A computer scientist's view of life, the universe, and everything. Foundations of Computer Science: Potential – Theory – Cognition, Lecture Notes in Computer Science, pages 201–208, Springer, 1997
  12. ↑ Brian Greene, Chapter 10 of: The Hidden Reality: Parallel Universes and the Deep Laws of the Cosmos, Knopf, 2011
  13. ↑ J. Schmidhuber. The Speed Prior: A New Simplicity Measure Yielding Near-Optimal Computable Predictions. Proceedings of the 15th Annual Conference on Computational Learning Theory (COLT 2002), Sydney, Australia, LNAI, 216–228, Springer, 2002
  14. ↑ J. Schmidhuber. Hierarchies of generalized Kolmogorov complexities and nonenumerable universal measures computable in the limit. International Journal of Foundations of Computer Science 13(4):587–612, 2002
  15. ↑ J. Schmidhuber. Ultimate Cognition à la Gödel. Cognitive Computation 1(2):177–193, 2009
  16. ↑ J. Schmidhuber. Optimal Ordered Problem Solver. Machine Learning, 54, 211–254, 2004
  17. ↑ J. Schmidhuber. Low-Complexity Art. Leonardo, Journal of the International Society for the Arts, Sciences, and Technology, 30(2):97–103, MIT Press, 1997
  18. ↑ http://robotics.idsia.ch/ The IDSIA Robotics Lab
  19. ↑ J. Leitner, S. Harding, P. Chandrashekhariah, M. Frank, A. Förster, J. Triesch and J. Schmidhuber. Learning Visual Object Detection and Localisation Using icVision. Biologically Inspired Cognitive Architectures, Vol. 5, 2013.
  20. ↑ J. Leitner, S. Harding, M. Frank, A. Förster and J. Schmidhuber. Humanoid Learns to Detect Its Own Hands. IEEE Congress on Evolutionary Computing (CEC), 2013.
  21. ↑ S. Harding, J. Leitner and J. Schmidhuber. Cartesian Genetic Programming for Image Processing (CGP-IP). Genetic Programming Theory and Practice X (Springer Tract on Genetic and Evolutionary Computation). pp 31-44. ISBN 978-1-4614-6845-5. Springer, Ann Arbor, 2013.
  22. ↑ http://www.aaaivideos.org/2013/ AAAI Video Competition 2013.
  23. ↑ M. Stollenga, L. Pape, M. Frank, J. Leitner, A. Förster and J. Schmidhuber. Task-Relevant Roadmaps: A Framework for Humanoid Motion Planning. IROS, 2013.

Sources

External links

This article is issued from Wikipedia - version of the Saturday, March 12, 2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.