Lectures by Marcus Hutter
Partial & Full Courses for Students
- Reinforcement Learning and Planning under Uncertainty
Winter Semester (2008) NICTA & ANU, Canberra, Lectures
- Introduction to Artificial Intelligence
Summer Semester (2007&2008) Australian National University, Canberra, Lectures
- Introduction to Statistical Machine Learning (more slides)
Summer Semester (2007&2008) NICTA & ANU, Canberra, Lectures
- On the Philosophical, Statistical, and Computational Foundations of Inductive Inference and Intelligent Agents
International Conference on Algorithmic Information Theory (2007), Sendai, Tutorial
- Combinatorics and Probability
Winter Semester (2006) Australian National University, Canberra, Lectures
- Universal Artificial Intelligence: Mathematical and Philosophical Foundations
Helsinki Graduate School in Computer Science and Engineering
(HeCSE 2006), Helsinki, Lectures
- How to Predict Sequences with Bayes, MDL, and Experts
International Conference on Machine Learning (2005), Bonn, Tutorial
- How to Predict with Bayes, MDL, and Experts
Machine Learning Summer School (2005), ANU/RSISE/NICTA Canberra, Lectures
- Theory Reading Group
Every Wednesday 15:00-16:30 (2004-2005), IDSIA, Organizer
- Algorithmic Information Theory and Machine Learning
Winter Semester (2003), University of Technology Munich, Lectures
- Quantum Electro Dynamics
Summer Semester (1995), LM-University Munich, Tutor
- Theoretical Mechanics
Winter Semester (1993), LM-University Munich, Tutor
Invited Lectures at Conferences & Workshops
-
The Fastest and Shortest Algorithm for All Well-Defined Problems
5th Turing Days Conference on Randomness and Complexity (2006), Bilgi University, Istanbul
- On the Foundations of Universal Sequence Prediction
Symposium on Theory and Applications of Models of Computation
(TAMC-2006), Learning Theory Session, Beijing
- Universal Artificial Intelligence
Swiss Mathematical Society,
Fall Meeting (SMS 2005), Lugano
- Theoretically Optimal Program Induction and Universal Artificial Intelligence
Inductive Programming Workshop W1 at
(ICML-2005), Bonn
- MDL Predictions based on Kolmogorov Complexity
Centennial Seminar on Kolmogorov Complexity and Applications (2003), Dagstuhl
- On the Existence and Convergence of Universal Priors
Workshop on Computability and Randomness (2003), Uni-Heidelberg
- Solomonoff Induction and the Foundations of Occam's, Epicurus', Bayes', and Utility Principles
Workshop on Foundations of Occam's razor (NIPS-2001), Vancouver
-
An effective Procedure for Speeding up Algorithms
Conference on Mathematical Approaches to Biological Computation (MaBiC-2001), Lavin
Workshop on Algorithmic Information Theory (TAI-2001), Porquerolles
- Universal Sequential Decisions in Unknown Environments
Workshop on Universal Learning Algorithms and Optimal Search (NIPS-2002), Vancouver
5th European Workshop on Reinforcement Learning (EWRL-2001), Utrecht
Lectures at Conferences
- The Loss Rank Principle for Model Selection
20th Annual Conf. on Learning Theory (COLT-2007), Sendai
- General Discounting versus Average Reward
16th International Conf. on Algorithmic Learning Theory (ALT-2006), Barcelona
- Universal Learning of Repeated Matrix Games
Annual Machine Learning Conference of Belgium and The Netherlands (Benelearn-2006), Ghent
- Fast Non-Parametric Bayesian Inference on Infinite Trees
15th International Conference on Artificial Intelligence and Statistics (AISTATS-2005), Barbados
- Universal Convergence of Semimeasures on Individual Random Sequences
15th International Conf. on Algorithmic Learning Theory (ALT-2004), Padova
Kolmogorov Complexity and Applications
(Dagstuhl-2006), Germany
- Prediction with Expert Advice by Following the Perturbed Leader for General Weights
15th International Conf. on Algorithmic Learning Theory (ALT-2004), Padova
- Online Prediction - Bayes versus Experts
EU PASCAL Workshop (LTBIP-2004), London
- Self-Optimizing and Pareto-Optimal Policies
in General Environments based on Bayes-Mixtures
15th Annual Conference on Computational Learning Theory (COLT-2002), Sydney
- Fitness Uniform Selection to Preserve Genetic Diversity
Congress on Evolutionary Computation (CEC-2002), Honolulu
Conference of the European Chapter on Combinatorial Optimization (ECCO-2002), Lugano
- Distribution of Mutual Information
14th Conference on Neural Information Processing Systems (NIPS-2001), Vancouver
- General Loss Bounds for Universal Sequence Prediction
18th International Conference on Machine Learning (ICML-2001), Williamstown
- Towards a Universal Theory of Artificial
Intelligence based on Algorithmic Probability and Sequential Decisions
12th European Conference on Machine Learning (ECML-2001), Freiburg
- Convergence and Error Bounds for Universal Prediction of Nonbinary Sequences
12th European Conference on Machine Learning (ECML-2001), Freiburg
(Invited) Lectures at Universities
- On Universal Induction and Intelligent Agents
Australian National University
(ANU 2008), Australia
- Bayes-Optimal Policies in General Environments
University of Alberta
(UA 2007), Edmonton
- On the Philosophical, Statistical, and Computational Foundation of Inductive Inference
University of Queensland (UQLD 2008), Brisbane
University of Alberta
(UA 2007), Edmonton
- On Universal Prediction and Bayesian Confirmation
Swiss Federal Institute of Technology Zurich
(ETHZ 2006), Zürich
- Bayesian PC-Regression for Detecting Aberrations in DNA of Cancer Cells
Swiss Federal Institute of Technology Zurich
(ETHZ 2006), Zürich
Dalle Molle Institute for Artificial Intelligence (IDSIA 2005), Lugano
- Bayesian and Universal Induction
Swiss Federal Institute of Technology Zurich (ETHZ 2006), Zürich
- Universal Prediction: Concepts, Tools and Applications
Oncology Institute of Southern Switzerland & Dalle Molle Institute for Artificial Intelligence (IOSI/IDSIA 2005), Lugano
- Foundations of Machine Learning = Information + Decision Theory
University of Alberta
(UA 2007), Edmonton
Australian National University (ANU 2005), Canberra
- Fast Non-Parametric Bayesian Inference on Infinite Trees
University of Sydney (USYD 2005), Sydney
- Optimal Sequential Decisions Based on Algorithmic Probability
Swiss Federal Institute of Technology Zurich (ETHZ 2004), Zürich
California Institute of Technology (CALTECH 2003), Pasadena (Los Angeles)
- Bayesian Mutual Information and Robust Feature Selection
Ludwig-Maximilian Univerity Munich (LMU 2004), Munich
- MDL Predictions based on Kolmogorov Complexity
Boston University (BU 2003), Boston
- Towards a Universal Theory of Artificial
Intelligence based on Algorithmic Probability and Sequential Decisions
Workshop on Universal Learning Algorithms and Optimal Search (NIPS-2002), Vancouver
University of Queensland (UQLD 2002), Brisbane
Monash University (2002), Melbourne
University of New South Wales (UNSW 2002), Sydney
Australian National University (ANU 2002), Canberra
Boston University (BU 2002), Boston
Centrum voor Wiskunde en Informatica (CWI 2002), Amsterdam
-
The Fastest and Shortest Algorithm for All Well-Defined Problems
University of New South Wales (UNSW 2005), Sydney
California Institute of Technology (CALTECH 2003), Pasadena (Los Angeles)
Centrum voor Wiskunde en Informatica (CWI 2001), Amsterdam
- New Error Bounds for Solomonoff Prediction
University of Technology Munich (TUM 2000), Munich
- A Theory of Universal Artificial Intelligence
based on Algorithmic Complexity
Istituto Dalle Molle di Studi sull'Intelligenza Artificiale (IDSIA 2000), Lugano
University of Technology (TUM 2000), Munich
- Instantons in QCD: Theory and Application of the
Instanton Liquid Model
University of Tel Aviv (1995), Tel Aviv
- Instantons and Meson Correlators in QCD
CERN (1995), Geneve
More Lectures