Everybody "knows" that neural networks need more than a single layer of nonlinear units to compute interesting functions. We show that this is false if one employs winner-take-all as nonlinear unit: • Any boolean function can be computed by a single k-winner-take-all unit applied to weighted sums of the input variables. • Any continuous function can be approximated arbitrarily well by a single soft winner-take-all unit applied to weighted sums of the input variables. • Only positive weights are needed in these (linear) weighted sums. This may be of interest from the point of view of neurophysiology, since only 15% of the synapses in the cortex are inhibitory. In addition it is widely believed that there are special microcircuits in the cortex that compute winner-take-all. • Our results support the view that winner-take-all is a very useful basic computational unit in Neural VLSI: □ it is wellknown that winner-take-all of n input variables can be computed very efficiently with 2n transistors (and a total wire length and area that is linear in n) in analog VLSI [Lazzaro et at., 1989] □ we show that winner-take-all is not just useful for special purpose computations, but may serve as the only nonlinear unit for neural circuits with universal computational power □ we show that any multi-layer perceptron needs quadratically in n many gates to compute winner-take-all for n input variables, hence winner-take-all provides a substantially more powerful computational unit than a perceptron (at about the same cost of implementation in analog VLSI). Complete proofs and further details to these results can be found in [Maass, 2000].
[1]
T. Kailath,et al.
Discrete Neural Computation: A Theoretical Foundation
,
1995
.
[2]
John Lazzaro,et al.
Winner-Take-All Networks of O(N) Complexity
,
1988,
NIPS.
[3]
Wolfgang Maass,et al.
On the Computational Power of Winner-Take-All
,
2000,
Neural Computation.
[4]
S. Grossberg.
Contour Enhancement , Short Term Memory , and Constancies in Reverberating Neural Networks
,
1973
.
[5]
Kiichi Urahama,et al.
K-winners-take-all circuit with O(N) complexity
,
1995,
IEEE Trans. Neural Networks.
[6]
J. Meador,et al.
Pulse Coded Winner-Take-All Networks
,
1994
.
[7]
Christof Koch,et al.
Analog VLSI Circuits for Attention-Based, Visual Tracking
,
1996,
NIPS.
[8]
Giacomo Indiveri,et al.
Modeling Selective Attention Using a Neuromorphic Analog VLSI Device
,
2000,
Neural Computation.
[9]
Timothy X. Brown,et al.
Neural network design for switching network control
,
1991
.