Semi-Supervised Methods for Handwritten Character Recognition using Active Learning (2007)
There are a number of supervised machine learning methods such as classiffers pretrained using restricted Boltzmann machines and convolutional networks that work very well for handwritten character...
Recurrent Neural Network Modeling of Nearshore Sandbar Behavior (2007)
Pape, L., Ruessink, B.G., Wiering, M.A., Turner, I.L.
The temporal evolution of nearshore sandbars (alongshore ridges of sand fringing coasts in water depths less than 10 m and of paramount importance for coastal safety) is commonly predicted using...
Computing Optimal Stationary Policies for Multi-objective Markov Decision Processes (2007)
This paper describes a novel algorithm called CONMODP for computing Pareto optimal policies for deterministic multi-objective sequential decision problems. CON-MODP is a value iteration based...
Reinforcement Learning in Continuous Action Spaces (2007)
Hasselt, H. Van, Wiering, M.A.
Quite some research has been done on Reinforcement Learning in continuous environments, but the research on problems where the actions can also be chosen from a continuous space is much more limited....
Convergence of Model-Based Temporal Difference Learning for Control (2007)
Hasselt, H. Van, Wiering, M.A.
A theoretical analysis of Model-Based Temporal Difference Learning for Control is given, leading to a proof of convergence. This work differs from earlier work on the convergence of Temporal...
Two Novel On-policy Reinforcement Learning Algorithms based on TD(lambda)-methods (2007)
Wiering, M.A., Hasselt, H. Van
This paper describes two novel on-policy reinforcement learning algorithms, named QV(lambda)-learning and the actor critic learning automaton (ACLA). Both algorithms learn a state value-function...
A model based method for automatic facial expression recognition (2006)
Kuilenburg, H. Van, Wiering, M.A., Uyl, M. Den
Automatic facial expression recognition is a research topic with interesting applications in the field of human-computer interaction, psychology and product marketing. The classification accuracy for...
Strategies for Ontology Negotiation : Finding the Right Level of Generality (2006)
Diggelen, J. Van, Jong, E.D. De, Wiering, M.A.
International workshop on agent communication 2006, held with AAMAS 2006 In heterogeneous multi agent systems, communication is hampered by the lack of shared ontologies. Ontology negotiation is a...
Cognitive Developmental Pattern Recognition : Learning to Learn (2006)
Zant, T. Van Der, Schomaker, L., Wiering, M.A., Brink, A.
It can be very difficult to create software systems which capture the knowledge of an expert. It is an expensive and laborious process that often results in a suboptimal solution. This article...
Strategies for ontology negotiation: Finding the right level of generality (2006)
Diggelen, J. Van, Wiering, M.A., Jong, E.D. De
International workshop on agent communication 2006, held with AAMAS 2006 In heterogeneous multi agent systems, communication is hampered by the lack of shared ontologies. Ontology negotiation is a...
Autonomous Intelligent Robots at Utrecht University (2004)
Wiering, M.A., Leijnen, S., Koster, A., Weers, S. Van, Back, Walter De
Model-Based Reinforcement (2003)
M. A. Wiering, R. P. Sal/ustowicz, J. Schmidhuber
Introduction Game playing programs have been a major focus of artificial intelligence (AI) research. How to represent and evaluate positions? How to use planning for exploiting evaluations to select...
A Characterization of Sapient Agents (2003)
Otterlo, M. Van, Wiering, M.A., Dastani, M.M., Meyer, J-J.Ch.
Post-Processing for MCMC (2003)
Jong, E.D. De, Wiering, M.A., Drugan, M.M.
Markov Chain Monte Carlo methods (MCMC) can sample from a target distribution and approximate this distribution. MCMC methods employ the apriori known unnormalized target distribution to decide on...
Hierarchical mixtures of naive Bayes classifiers (2002)
Naive Bayes classifiers tend to perform very well on a large number of problem domains, although their representation power is quite limited compared to more sophisticated machine learning...
Clockwork Orange: The Dutch RoboSoccer Team (2002)
Spaan, M.T.J., Wiering, M.A., Bartelds, R., Donkervoort, R., Jonker, P., Groen, F.C.A.
Proceedings of the 12th Belgian-Dutch Conference on Machine Learning (2002)
The Twelfth Belgian-Dutch Conference on Machine Learning (Benelearn'02) has gathered a wide variety of researchers interested in many different topics in machine learning (ML). Decision trees,...
Proceedings of the Fifth European workshop on Reinforcement Learning (2001)
The Fifth European Workshop on Reinforcement Learning (EWRL-5) has gathered a wide variety of researchers interested in many different topics in reinforcement learning (RL). First of all, several...
The influence of Communication on the Choice to Behave Cooperativly (2001)
In this paper we investigate the learning of cooperation and communication in a multi agent system. A predator prey pursuit domain is defined in which predators can learn to both non-cooperatively...
Clockwork Orange: The Dutch RoboSoccer Team (2001)
Spaan, M (stud), Wiering, M.A., Bartelds, R., Donkervoort, R., Jonker, P., Groen, F.C.A.
Speeding up Q (λ)- learning (1998)
Wiering, M.A., Schmidhuber, J.
Q(λ)learning uses TD(λ)methods to accelerate Q-learning. The worst case complexity for a single update step of previous online Q(λ) implementations based on lookup tables is bounded by the size of...
Learning to Control Forest Fires (1998)
Forest fires are an important environmental problem. This paper describes a methodology for constructing an intelligent system which aims to support the human expert's decision making in fire...
Learning Team Strategies: Soccer Case Studies (1998)
Wiering, M.A., Salustowicz, R.P., Schmidhuber, J.
We use simulated soccer to study multiagent learning. Each team's players (agents) share action set and policy, but may behave differently due to position-dependent inputs. All agents making up a...
Efficient Model-Based Exploration (1998)
Wiering, M.A., Schmidhuber, J.
Model-Based Reinforcement Learning (MBRL) can greatly profit from using world models for estimating the consequences of selecting particular actions: an animat can construct such a model from its...
Wiering, M.A., Schmidhuber, J.
Q(lambda)-learning uses TD(lambda)-methods to accelerate Q-learning. The update complexity of previous online Q(lambda)implementations based on lookup-tables is bounded by the size of the...
Learning Exploration Policies with Models (1998)
Wiering, M.A., Schmidhuber, J.
Reinforcement learning can greatly profit from world models updated by experience and used for computing policies. Fast discovery of near optimal policies however requires to focus on "useful"...
Learning to Control Forest Fires (1998)
Forest fires are an important environmental problem. This paper describes a methodology for constructing an intelligent system which aims to support the human expert's decision making in fire...
Speeding Up Q(lambda)-learning (1998)
Wiering, M.A., Schmidhuber, J.
Q(lambda)-learning uses TD(lambda)-methods to accelerate Q learning. The worst case complexity for a single update step of previous online Q(lambda)implementations based on lookup tables is bounded...
CIREC: Cluster Correlogram Image Retrieval and Categorization using MPEG-7 Descriptors (1998)
Content-based image retrieval is generally about understanding of information in the images concerned. The more the system is able to understand the content of images the more effective it will be in...
Evolving Soccer Strategies (1997)
Salustowicz, R., Wiering, M.A., Schmidhuber, J.
We study multiagent learning in a simulated soccer scenario. Players from the same team share a common policy for mapping inputs to actions. They get rewarded or punished collectively in case of...
Learning Team Strategies With Multiple Policy-Sharing Agents: A Soccer Case Study (1997)
Salustowicz, R., Wiering, M.A., Schmidhuber, J.
We use simulated soccer to study multiagent learning. Each team's players (agents) share action set and policy, but may behave differently due to position-dependent inputs. All agents making up a...
On learning soccer strategies (1997)
Salustowicz, R., Wiering, M.A., Schmidhuber, J.
We use simulated soccer to study multiagent learning. Each team's players (agents) share action set and policy but may behave differently due to position-dependent inputs. All agents making up a team...
Schmidhuber, J., Zhao, J., Wiering, M.A.
We study task sequences that allow for speeding up the learners average reward intake through appropriate shifts of inductive bias changes of the learner's policy. To evaluate long-term effects of...