Markov Chain Neural Networks
This work addresses the problem of incorporating stochastic processes into neural networks for researchers in machine learning and simulation domains, but it appears incremental as it modifies existing neural network models.
The authors introduced a modified neural network model capable of simulating Markov Chains, demonstrating applications such as a random walker for Brownian motion simulation and a non-deterministic Tic-Tac-Toe network.
In this work we present a modified neural network model which is capable to simulate Markov Chains. We show how to express and train such a network, how to ensure given statistical properties reflected in the training data and we demonstrate several applications where the network produces non-deterministic outcomes. One example is a random walker model, e.g. useful for simulation of Brownian motions or a natural Tic-Tac-Toe network which ensures non-deterministic game behavior.