Deep Learning & Computer Vision with Fast Deep Neural Nets. The future of search engines and robotics lies in image and video recognition. Since 2009, our Deep Learning team has won 9 (nine) first prizes in important and highly competitive international contests (with secret test sets known only to the organisers), far more than any other team. Our neural nets also set numerous world records, and were the first Deep Learners to win pattern recognition contests in general (2009), the first to win object detection contests (2012), the first to win a pure image segmentation contest (2012), and the first machine learning methods to reach superhuman visual recognition performance in a contest (2011). Compare this Google Tech Talk (2011) and JS' first Deep Learning system of 1991, with a Deep Learning timeline 1962-2013. See also the history of computer vision contests won by deep CNNs on GPU since 2011. And check out the amazing Highway Networks (2015), the deepest of them all.

Gödel machine: An old dream of computer scientists is to build an optimally efficient universal problem solver. The Gödel machine can be implemented on a traditional computer and solves any given computational problem in an optimal fashion inspired by Kurt Gödel's celebrated self-referential formulas (1931). It starts with an axiomatic description of itself, and we may plug in any utility function, such as the expected future reward of a robot. Using an efficient proof searcher, the Gödel machine will rewrite any part of its software (including the proof searcher) as soon as it has found a proof that this will improve its future performance, given the utility function and the typically limited computational resources. Self-rewrites are globally optimal (no local maxima!) since provably none of all the alternative rewrites and proofs (those that could be found by continuing the proof search) are worth waiting for. The Gödel machine formalizes I. J. Good's informal remarks (1965) on an "intelligence explosion" through self-improving "super-intelligences". Summary. FAQ.

Optimal Ordered Problem Solver. OOPS solves one task after another, through search for solution- computing programs. The incremental method optimally exploits solutions to earlier tasks when possible - compare principles of Levin's optimal universal search. OOPS can temporarily rewrite its own search procedure, efficiently searching for faster search methods (metasearching or metalearning). It is applicable to problems of optimization or prediction. Talk slides.

Super Omegas and Generalized Kolmogorov Complexity and Algorithmic Probability. Kolmogorov's (left) complexity K(x) of a bitstring x is the length of the shortest program that computes x and halts. Solomonoff's algorithmic probability of x is the probability of guessing a program for x. Chaitin's Omega is the halting probability of a Turing machine with random input (Omega is known as the "number of wisdom" because it compactly encodes all mathematical truth). Schmidhuber generalized all of this to non-halting but converging programs. This led to the shortest possible formal descriptions and to non-enumerable but limit-computable measures and Super Omegas, and even has consequences for computable universes and optimal inductive inference. Slides.

Universal Learning Algorithms. There is a theoretically optimal way of predicting the future, given the past. It can be used to define an optimal (though noncomputable) rational agent that maximizes its expected reward in almost arbitrary environments sampled from computable probability distributions. This work represents the first mathematically sound theory of universal artificial intelligence - most previous work on AI was either heuristic or very limited.

Speed Prior. Occam's Razor: prefer simple solutions to complex ones. But what exactly does "simple" mean? According to tradition something is simple if it has a short description or program, that is, it has low Kolmogorov complexity. This leads to Solomonoff's & Levin's miraculous probability measure which yields optimal though noncomputable predictions, given past observations. The Speed Prior is different though: it is a new simplicity measure based on the fastest way of describing objects, not the shortest. Unlike the traditional one, it leads to near-optimal computable predictions, and provokes unusual prophecies concerning the future of our universe. Talk slides. Transcript of TEDx talk.

In the Beginning was the Code. In 1996 Schmidhuber wrote the first paper about all possible computable universes. His `Great Programmer' is consistent with Zuse's thesis (1967) of computable physics, against which there is no physical evidence, contrary to common belief. If everything is computable, then which exactly is our universe's program? It turns out that the simplest program computes all universes, not just ours. Later work (2000) on Algorithmic Theories of Everything analyzed all the universes with limit-computable probabilities as well as the very limits of formal describability. This paper led to above-mentioned generalizations of algorithmic information and probability and Super Omegas as well as the Speed Prior. See comments on Wolfram's 2002 book and letter on randomness in physics (Nature 439, 2006). Talk slides, TEDx video, transcript.

Learning Robots. Some hardwired robots achieve impressive feats. But they do not learn like babies do. Traditional reinforcement learning algorithms are limited to simple reactive behavior and do not work well for realistic robots. Hence robot learning requires novel methods for learning to identify important past events and memorize them until needed. Our group is focusing on the above-mentioned recurrent neural networks, RNN evolution, Compressed Network Search, and policy gradients. Collaborations: with UniBW on robot cars, with TUM-AM on humanoids learning to walk, with DLR on artificial hands. New IDSIA projects on developmental robotics with curious adaptive humanoids have started in 2009. See AAAI 2013 Best Student Video.

Financial Forecasting. Our most lucrative neural network application employs a second-order method for finding the simplest model of stock market training data.

Interestingness & Active Exploration & Artificial Curiosity & Theory of Surprise (1990-2010). Schmidhuber's curious learning agents like to go where they expect to learn something. These rudimentary artificial scientists or artists are driven by intrinsic motivation, losing interest in both predictable and unpredictable things. A basis for much of the recent work in Developmental Robotics since 2004. According to Schmidhuber's formal theory of creativity, art and science and humor are just by-products of the desire to create / discover more data that is predictable or compressible in hitherto unknown ways! See Singularity Summit talk (2009).

Reinforcement Learning in partially observable worlds. Just like humans, reinforcement learners are supposed to maximize expected pleasure and minimize expected pain. Most traditional work is limited to reactive mappings from sensory inputs to actions. Our approaches (1989-2003) for partially observable environments are more general: they learn how to use memory and internal states, sometimes through evolution of RNN. The first universal reinforcement learner is optimal if we ignore computation time, and here is one that is optimal if we don't. The novel Natural Evolution Strategies (2008-) link policy gradients to evolution. See also Compressed Network Search.

Unsupervised learning; non-linear ICA; history compression. Pattern recognition works better on non-redundant data with independent components. Schmidhuber's Predictability Minimization (1992) was the first non-linear neural algorithm for learning to encode redundant inputs in this way. It is based on co-evolution of unsupervised adversarial neural networks that fight each other in a minimax game. His neural history compressors (1991) compactly encode sequential data for Deep Learning. And Lococode unifies regularization and unsupervised learning. The feature detectors generated by such unsupervised methods resemble those of our more recent supervised neural computer vision systems.

Metalearning Machines / Learning to Learn / Recursive Self- Improvement. Can we construct metalearning algorithms that learn better learning algorithms? This question has been a main drive of Schmidhuber's research since his 1987 diploma thesis. In 1993 he introduced self-referential weight matrices, and in 1994 self-modifying policies trained by the "success-story algorithm" (talk slides). His first bias-optimal metalearner was the above-mentioned Optimal Ordered Problem Solver (2002), and the ultimate metalearner is the Gödel machine (2003). See also work on "learning to think."

Automatic Subgoal Generators and Hierarchical Learning. There is no teacher providing useful intermediate subgoals for our reinforcement learning systems. In the early 1990s Schmidhuber introduced gradient-based (pictures) adaptive subgoal generators; in 1997 also discrete ones.

Program Evolution and Genetic Programming. As an undergrad Schmidhuber used Genetic Algorithms to evolve computer programs on a Symbolics LISP machine at SIEMENS AG. Two years later this was still novel: In 1987 he published world's 2nd paper on pure Genetic Programming (the first was Cramer's in 1985) and the first paper on Meta-Genetic Programming.

Learning Economies with Credit Conservation. In the late 1980s Schmidhuber developed the first credit-conserving reinforcement learning system based on market principles, and also the first neural one.

Neural Heat Exchanger. Like a physical heat exchanger, but with neurons instead of liquid. Perceptions warm up, expectations cool down.

Fast weights instead of recurrent nets. A slowly changing feedforward neural net learns to quickly manipulate short-term memory in quickly changing synapses of another net. More on fast weights. Evolution of fast weight control.

Complexity-Based Theory of Beauty. In 1997 Schmidhuber claimed: among several patterns classified as "comparable" by some subjective observer, the subjectively most beautiful is the one with the simplest description, given the observer's particular method for encoding and memorizing it. Exemplary applications include low-complexity faces and Low-Complexity Art, the computer-age equivalent of minimal art (Leonardo, 1997). A low-complexity artwork such as this Femme Fractale both `looks right' and is computable by a short program; a typical observer should be able to see its simplicity. The drive to create such art is explained by the formal theory of creativity.

Artificial Ants & Swarm Intelligence. IDSIA's Artificial Ant Algorithms are multiagent optimizers that use local search techniques and communicate via artificial pheromones that evaporate over time. They broke several important benchmark world records. This work got numerous reviews in journals such as Nature, Science, Scientific American, TIME, NY Times, Spiegel, Economist, etc. It led to an IDSIA spin-off company called ANTOPTIMA. See also the AAAI 2011 best video on swarmbots.

All cartoons & artwork & Fibonacci web design templates copyright © by Jürgen Schmidhuber (except when indicated otherwise).

Image below taken from an interview for Deutsche Bank during the 2019 World Economic Forum in Davos.