Possible hardware-based solutions leading to cognitive computing


The hardware emulation of stochastic invertible logic based on Arduino microcontrollers or implemented in magnetic tunnel junctions, the hardware belief networks, have been proposed recently by researchers from the School of Electrical and Computer Engineering, Purdue University (IN), USA. Importantly, the provided solutions realize not only standard stochastic computing of the Boolean type, but do it very precisely and in the invertible manner. The proposed computing network imitating synapses consists of elementary switches based on stochastic magnetic tunnel junctions. The junctions can provide not only a unique output in response to a set of inputs (what is classical effect nowadays) but, first of all, can provide all the inputs that are consistent with a given output. This fact is of fundamental importance since in real neurons an elementary biological cell has the ability to modify number of local connections with the synapse, in order to enhance or reduce the coupling efficiency of signal transmission. More details can be found in the following papers and the related references:  A. Z. Pervaiz, L. A. Ghantasala, K. Y. Camsari, and S. Datta, Hardware emulation of stochastic p-bits for invertible logic, Scientific Reports 7, 10994 (2017), and in K.Y. Camsari, R. Faria, B. M. Sutton and S. Datta, Stochastic p-bits for Invertible Boolean Logic, Phys. Rev. X, 3, 031014 (2017).