Learning in the presence of concept drift and hidden contexts

On-line learning in domains where the target concept depends on some hidden context poses serious problems. A changing context can induce changes in the target concepts, producing what is known as concept drift. We describe a family of learning algorithms that flexibly react to concept drift and can take advantage of situations where contexts reappear. The general approach underlying all these algorithms consists of (1) keeping only a window of currently trusted examples and hypotheses; (2) storing concept descriptions and reusing them when a previous context re-appears; and (3) controlling both of these functions by a heuristic that constantly monitors the system's behavior. The paper reports on experiments that test the systems' perfomance under various conditions such as different levels of noise and different extent and rate of concept drift.

[1]  Ryszard S. Michalski,et al.  A Theory and Methodology of Inductive Learning , 1983, Artificial Intelligence.

[2]  Leslie G. Valiant,et al.  A theory of the learnable , 1984, STOC '84.

[3]  Stephen Grossberg,et al.  Competitive Learning: From Interactive Activation to Adaptive Resonance , 1987, Cogn. Sci..

[4]  Michael Lebowitz,et al.  Experiments with incremental concept formation: UNIMEM , 2004, Machine Learning.

[5]  Pat Langley,et al.  Hill-Climbing Theories of Learning , 1987 .

[6]  Allen Newell,et al.  Some Chunks Are Expensive , 1988, ML.

[7]  Shaul Markovitch,et al.  The Role of Forgetting in Learning , 1988, ML.

[8]  Dana Angluin,et al.  Queries and concept learning , 1988, Machine Learning.

[9]  Miroslav Kubat.  Floating approximation in time-varying knowledge bases , 1989, Pattern Recognit. Lett..

[10]  David Haussler,et al.  Learnability and the Vapnik-Chervonenkis dimension , 1989, JACM.

[11]  Ronald L. Rivest,et al.  Learning Time-Varying Concepts , 1990, NIPS.

[12]  Steven Minton,et al.  Quantitative Results Concerning the Utility of Explanation-based Learning , 1988, Artif. Intell..

[13]  Philip M. Long,et al.  Tracking drifting concepts using random examples , 1991, Annual Conference Computational Learning Theory.

[14]  Andrew W. Moore,et al.  Fast, Robust Adaptive Control by Learning only Forward Models , 1991, NIPS.

[15]  Wolfgang Maass,et al.  On-line learning with an oblivious environment and the power of randomization , 1991, COLT '91.

[16]  Ronald L. Rivest,et al.  Incrementally Learning Time-Varying Half Planes , 1991, NIPS.

[17]  Miroslav Kubat,et al.  FAVORIT: Concept formation with ageing of knowledge , 1992, Pattern Recognit. Lett..

[18]  Philip M. Long,et al.  Apple tasting and nearly one-sided learning , 1992, Proceedings., 33rd Annual Symposium on Foundations of Computer Science.

[19]  Miroslav Kubat.  A machine learning-based approach to load balancing in computer networks , 1992 .

[20]  Gerhard Widmer,et al.  Learning Flexible Concepts from Streams of Examples: FLORA 2 , 1992, ECAI.

[21]  Miroslav Kubat.  Conceptual Inductive Learning: The Case of Unreliable Teachers , 1992, Artif. Intell..

[22]  Marcos Salganicoff,et al.  Density-Adaptive Learning and Forgetting , 1993, ICML.

[23]  Pat Langley,et al.  Average-Case Analysis of a Nearest Neighbor Algorithm , 1993, IJCAI.

[24]  Gerhard Widmer,et al.  Effective Learning in Dynamic Environments by Explicit Context Tracking , 1993, ECML.

[25]  Fredrik Kilander,et al.  COBBIT - A Control Procedure for COBWEB in the Presence of Concept Drift , 1993, ECML.

[26]  Miroslav Kubat,et al.  Flexible concept learning in real-time systems , 1993, J. Intell. Robotic Syst..

[27]  Marcos Salganicoff,et al.  Explicit Forgetting Algorithms for Memory Based Learning , 1993 .

[28]  Gerhard Widmer.  Combining Robustness and Flexibility in Learning Drifting Concepts , 1994, ECAI.

[29]  Peter D. Turney.  Robust Classification with Context-Sensitive Features , 2002, ArXiv.

[30]  J. C. Schlimmer,et al.  Incremental learning from noisy data , 2004, Machine Learning.

[31]  Tom M. Mitchell,et al.  Explanation-Based Generalization: A Unifying View , 1986, Machine Learning.

[32]  Philip M. Long,et al.  Tracking drifting concepts by minimizing disagreements , 2004, Machine Learning.

[33]  Steven Salzberg,et al.  A Weighted Nearest Neighbor Algorithm for Learning with Symbolic Features , 2004, Machine Learning.

[34]  D. Kibler,et al.  Instance-based learning algorithms , 2004, Machine Learning.

[35]  Leslie Pack Kaelbling,et al.  Associative Reinforcement Learning: Functions in k-DNF , 1994, Machine Learning.

[36]  Steven Salzberg,et al.  A Nearest Hyperrectangle Learning Method , 1991, Machine Learning.