Machine Learning is an international forum for research on computational approaches to learning. The journal publishes articles reporting substantive results on a wide range of learning methods applied to a variety of task domains, including but not limited to:

Methods: Supervised and unsupervised learning methods (including learning decision and regression trees, rules, connectionist networks, probabilistic networks and other statistical models, genetic algorithms, genetic programming, inductive logic programming, case-based methods); Reinforcement learning; Explanation-based learning; Analogical learning methods; Automated knowledge acquisition; Learning from instruction; Visualization of patterns in data; Learning in integrated architectures; Multistrategy learning; Multi-agent learning.

Task Domains: Classification, regression, recognition, and prediction; Problem solving and planning; Reasoning and inference; Data mining; Scientific discovery; Information Retrieval; Natural language processing; Design and diagnosis; Vision and speech perception; Robotics and control; Combinatorial optimization; Industrial, financial, and scientific applications of all kinds.

Papers may describe exploratory research, research on problems and methods, and applications research. Exploratory research papers identify new learning tasks and problems and motivate their importance. Papers analyzing learning problems and methods make claims about learning problems (e.g., inherent complexity) or methods (e.g., relative performance of alternative algorithms) supported by empirical studies, theoretical analysis, or comparison to psychological phenomena. Applications papers show how to apply learning methods to solve important applications problems. All papers must describe their work in ways that can be verified or replicated by other researchers. All papers must discuss knowledge representation and performance task assumptions as well as describing the learning component, and all papers should relate their contribution to other work in machine learning. Variations from these prototypes, such as critical reviews of existing work, will be considered provided they make a clear contribution to the field.

This page maintained by Robert C. Holte holte@cs.ualberta.ca