2:00 PM - 2:30 PM
[12p-B414-3] Cellular Automata Reservoir Computing Acceleration
Keywords:reservoir computing, cellular automata
Neural networks based on deep learning require a large amount of computational cost in return for high inference accuracy. For reducing training and implementation costs, reservoir computing substitutes a parameter-fixed recurrent neural network for the hidden layers that occupy the majority of computational cost in a neural network. This lecture introduces a reservoir computing method using cellular automaton and its implementation.