An ASIC Analog Neuromorphic Computer for Evolvable Hardware Applications

Aims: This paper presents a design for a custom Application-Specific-Integrated-Circuit (ASIC) VLSI continuous time recurrent neural network computer suitable for use in Evolvable Hardware (EH) applications. Study Design: Extensive testing of a fabricated device will be used to demonstrate that the designed and fabricated neural chip possesses excellent behavioral congruence to the differential equation and ASIC hardware forms of neural networks programmed into the chip. Place and Duration of Study: Department of Computer Science and Engineering, Wright State University, between 2009 and 2012. Methodology: The presented ASIC neural chip has been designed with specific concentration on the CMOS sub-threshold design concepts. This CMOS sub-threshold design forms the basis for underlying neural computation and also the current-mode Digital-to-Analog Converter (DAC) that can be used to program neuron configurations. The proposed designed has been developed to be immune to any faults introduced thru fabrication, at least to the extent that is non-detrimental to underlying neural behavior. Results: Ten separate intrinsic CTRNN learning runs were conducted on the fabricated chips. Each test was conducted on a separate fabricated chip to assess intrinsic to extrinsic transferability across individual instantiations of the device. Further, as mentioned earlier, a secondary set of tests were conducted that involved performing intrinsic match analysis for 15 (separate) extrinsically learnt CTRNN configurations to test extrinsic to intrinsic transferability. Based on the comparison metrics computed between the simulated and the fabricated chip, it has been demonstrated that the observed worst case average mismatch across all computed outputs of the four neuron CTRNNs is about seven percent on amplitude with near perfect matching for slope and frequency. Conclusion: Extensive testing of a fabricated device has been used to demonstrate that the analog computer possesses excellent behavioral congruence to the differential equation and ASIC hardware forms of neural networks are programmed into the chip. The major advantage of choosing the proposed CTRNNs chip for EH applications is that one can easily transition between model and circuit form no matter how the circuit was evolved. In this paper, we Original Research Article British Journal of Mathematics & Computer Science 4(7), 989-1003, 2014 990 demonstrated quite clearly that the barrier is either non-existent or very slight by having designed, fabricated, and tested an actual VLSI chip in the application that one would expect to be most difficult -evolution in hardware and modeling in differential equation form.

[1]  J. J. Hopfield,et al.  “Neural” computation of decisions in optimization problems , 1985, Biological Cybernetics.

[2]  Sanjay K. Boddhu,et al.  A Commercial Off-the-Shelf Implementation of an Analog Neural Computer , 2008, Int. J. Artif. Intell. Tools.

[3]  D. E. Goldberg,et al.  Genetic Algorithms in Search , 1989 .

[4]  David B. Fogel,et al.  System Identification Through Simulated Evolution: A Machine Learning Approach to Modeling , 1991 .

[5]  Robert G. Meyer,et al.  Analysis and Design of Analog Integrated Circuits , 1993 .

[6]  John C. Gallagher,et al.  Evolving neuromorphic flight control for a flapping-wing mechanical insect , 2010, Int. J. Intell. Comput. Cybern..

[7]  Garrison W. Greenwood,et al.  Introduction to Evolvable Hardware - A Practical Guide for Designing Self-Adaptive Systems , 2006 .

[8]  Sanjay K. Boddhu,et al.  Qualitative Functional Decomposition Analysis of Evolved Neuromorphic Flight Controllers , 2012, Appl. Comput. Intell. Soft Comput..

[9]  Thomas Bäck,et al.  Evolutionary computation: comments on the history and current state , 1997, IEEE Trans. Evol. Comput..

[10]  John C. Gallagher,et al.  An analysis of the search performance of a mini-population evolutionary algorithm for a robot-locomotion control problem , 2005, 2005 IEEE Congress on Evolutionary Computation.

[11]  J J Hopfield,et al.  Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

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

[13]  John C. Gallagher,et al.  A space saving digital VLSI evolutionary engine for CTRNN-EH devices , 2005, 2005 IEEE Congress on Evolutionary Computation.

[14]  Kumar Yelamarthi,et al.  Process Variation Aware Timing Optimization through Transistor Sizing in Dynamic CMOS Logic , 2008, ISQED 2008.

[15]  W. Guggenbuhl,et al.  On charge injection in analog MOS switches and dummy switch compensation techniques , 1990 .

[16]  Robert J. Wood,et al.  The First Takeoff of a Biologically Inspired At-Scale Robotic Insect , 2008, IEEE Transactions on Robotics.

[17]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[18]  Jack L. Meador,et al.  Programmable impulse neural circuits , 1991, IEEE Trans. Neural Networks.