论文信息 - A Local Algorithm to Learn Trajectories with Stochastic Neural Networks

A Local Algorithm to Learn Trajectories with Stochastic Neural Networks

This paper presents a simple algorithm to learn trajectories with a continuous time, continuous activation version of the Boltzmann machine. The algorithm takes advantage of intrinsic Brownian noise in the network to easily compute gradients using entirely local computations. The algorithm may be ideal for parallel hardware implementations.

Javier R. Movellan | J. Movellan

[1] N. Bershad,et al. Random differential equations in science and engineering , 1975, Proceedings of the IEEE.

[2] T. T. Soong,et al. Random differential equations in science and engineering , 1974 .

[3] Radford M. Neal. Asymmetric Parallel Boltzmann Machines are Belief Networks , 1992, Neural Computation.

[4] Javier R. Movellan,et al. Learning Continuous Probability Distributions with Symmetric Diffusion Networks , 1993, Cogn. Sci..

[5] Bruno Apolloni,et al. Learning by Asymmetric Parallel Boltzmann Machines , 1991, Neural Computation.