(

**Jürgen Schmidhuber ^{1}
Department of Computer Science
University of Colorado
Boulder, CO 80309, USA
Daniel Prelinger
Institut für Informatik
Technische Universität München
**

**
**

Prediction problems are among the most common learning problems for neural
networks (e.g. in the context of time series prediction, control, etc.).
With many such problems, however, perfect prediction is inherently
impossible. For such cases we present novel unsupervised systems that learn
to *classify* patterns such that the classifications are predictable while
still being as specific as possible. The approach can be related to
the IMAX method of Hinton, Becker and Zemel (1989, 1991). Experiments include
a binary stereo task proposed by
Becker and Hinton, which can be solved more readily by our
system.

- MOTIVATION AND BASIC APPROACH
- ALTERNATIVE DEFINITIONS OF

- RELATION TO PREVIOUS WORK
- ILLUSTRATIVE EXPERIMENTS
- FINDING PREDICTABLE LOCAL CLASS REPRESENTATIONS
- STEREO TASK
- FINDING PREDICTABLE DISTRIBUTED REPRESENTATIONS

- CONCLUSION
- ACKNOWLEDGEMENTS
- Bibliography
- About this document ...

Juergen Schmidhuber 2003-02-13