Perceiving without Learning: From Spirals to Inside/Outside Relations

Part of Advances in Neural Information Processing Systems 11 (NIPS 1998)

Bibtex Metadata Paper


Ke Chen, DeLiang Wang


As a benchmark task, the spiral problem is well known in neural net(cid:173) works. Unlike previous work that emphasizes learning, we approach the problem from a generic perspective that does not involve learning. We point out that the spiral problem is intrinsically connected to the in(cid:173) side/outside problem. A generic solution to both problems is proposed based on oscillatory correlation using a time delay network. Our simu(cid:173) lation results are qualitatively consistent with human performance, and we interpret human limitations in terms of synchrony and time delays, both biologically plausible. As a special case, our network without time delays can always distinguish these figures regardless of shape, position, size, and orientation.