On physical models of neural computation and their analog VLSI implementation

Examines computation in a framework where the problem is essentially that of extracting a signal from noise, filtering (selective amplification) or estimation. The discussion is relevant to computational tasks in sensory communication, such as vision, speech and natural language processing. We consider "real" systems, both natural (neural systems) and human engineered (silicon integrated circuits), where information processing takes the form of an irreversible physical process. We argue, and demonstrate experimentally, that it is possible to see the emergence of truly complex processing structures that are commensurate with the physical properties of the computational substrate and are therefore energetically efficient.<<ETX>>

[1]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[2]  Teuvo Kohonen,et al.  Self-Organization and Associative Memory, Second Edition , 1988, Springer Series in Information Sciences.

[3]  Zhaoping Li,et al.  Toward a Theory of the Striate Cortex , 1994, Neural Computation.

[4]  D. Puro.  The Retina. An Approachable Part of the Brain , 1988 .

[5]  Stephen E. Levinson,et al.  Adaptive acquisition of language , 1991 .

[6]  Christof Koch,et al.  Seeing Chips: Analog VLSI Circuits for Computer Vision , 1989, Neural Computation.

[7]  Robert W. Keyes,et al.  The physics of VLSI systems , 1987, Microelectronics systems design series.

[8]  J. Gunn,et al.  Thermodynamics of Nonlinearity and Noise in Diodes , 1968 .

[9]  Andreas G. Andreou,et al.  Characterization of subthreshold MOS mismatch in transistors for VLSI systems , 1994, J. VLSI Signal Process..

[10]  Lynn Conway,et al.  Introduction to VLSI systems , 1978 .

[11]  Carver Mead,et al.  Analog VLSI and neural systems , 1989 .

[12]  J. Fellrath,et al.  CMOS analog integrated circuits based on weak inversion operations , 1977 .

[13]  Teuvo Kohonen,et al.  Self-Organization and Associative Memory , 1988 .

[14]  Aleksandra Pavasović,et al.  Characterization of subthreshold MOS mismatch in transistors for VLSI systems , 1994 .

[15]  Andreas G. Andreou,et al.  A Contrast Sensitive Silicon Retina with Reciprocal Synapses , 1991, NIPS.

[16]  B. Gilbert.  A monolithic 16-channels analog array normalizer , 1984, IEEE Journal of Solid-State Circuits.

[17]  Tomaso Poggio,et al.  Computational vision and regularization theory , 1985, Nature.

[18]  Ralph Linsker,et al.  Self-organization in a perceptual network , 1988, Computer.

[19]  Hillis Wd,et al.  Parallel Scientific Computation , 1993 .

[20]  R. Douglas,et al.  A silicon neuron , 1991, Nature.

[21]  Tomaso Poggio,et al.  From Understanding Computation to Understanding Neural Circuitry , 1976 .

[22]  W. G. Owen,et al.  Temporal filtering in retinal bipolar cells. Elements of an optimal computation? , 1990, Biophysical journal.

[23]  Carver A. Mead,et al.  Neuromorphic electronic systems , 1990, Proc. IEEE.

[24]  B. Gilbert.  Translinear circuits: a proposed classification , 1975 .

[25]  R. Landauer.  Information is physical , 1991 .