Parameter Optimization of Winner-Take-All Circuit for Attention Shift Using Drosophila Food-Search Optimization Algorithm

The optimization of complementary metal oxide semiconductor (CMOS) winner-take-all (WTA) circuit that performs the selection process for attentive visual processing is a challenging problem in the present scenario. The CMOS WTA circuit uses excitatory feedback in a winner-take-all computation to produce a hysteresis in the selection from one location to the next. So, in this paper, the circuit parameter has been optimized using a newly developed algorithm called Drosophila food search optimization (DFO) algorithm, where an objective function is evaluated and minimized for attention shift. The circuit parameters are extracted from optimized objective function, and realization is done in CADENCE software using GPDK.

[1]  John Lazzaro,et al.  Winner-Take-All Networks of O(N) Complexity , 1988, NIPS.

[2]  Olivier Temam,et al.  Hardware spiking neurons design: Analog or digital? , 2012, The 2012 International Joint Conference on Neural Networks (IJCNN).

[3]  Giacomo Indiveri,et al.  A Current-Mode Hysteretic Winner-take-all Network, with Excitatory and Inhibitory Coupling , 2001 .

[4]  S Ullman,et al.  Shifts in selective visual attention: towards the underlying neural circuitry. , 1985, Human neurobiology.

[5]  Narayan Srinivasa,et al.  Energy-Efficient Neuron, Synapse and STDP Integrated Circuits , 2012, IEEE Transactions on Biomedical Circuits and Systems.

[6]  Kedar Nath Das,et al.  Drosophila Food-Search Optimization , 2014, Appl. Math. Comput..