Realization of reinforcement learning using multi-winners KFM associative memory | IEEE Conference Publication | IEEE Xplore