Toward Learning Systems That Integrate Different Strategies and Representations

An understanding of learning { the process by which a learner acquires and re nes a broad range of knowledge and skills { is central to the enterprise of building truly adaptive, exible, robust, and creative intelligent systems. Signi cant theoretical and empirical contributions to the characterization of learning in computational terms have emerged from research in a number of disparate research paradigms. The limitations of individual paradigms and of particular classes of techniques within each paradigm are beginning to be recognized. Converging lines of evidence from multiple sources, both theoretical as well as empirical, suggest that arti cial intelligence systems, in order to be able to deal with complex tasks such as recognizing and describing 3-dimensional objects, or communicating in natural language, must be able to e ectively utilize a range of learning algorithms operating with an adequate repertoire of representational structures. This paper draws on a broad range of research on learning in arti cial intelligence, connectionist networks, statistical and syntactic methods in pattern recognition, and evolutionary models to identify the similarities and di erences among, strengths and limitations of, and promising areas for cross-fertilization between, the di erent paradigms.