Nature Communications (Jul 2017)

Efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback

  • A. Emin Orhan,
  • Wei Ji Ma

DOI
https://doi.org/10.1038/s41467-017-00181-8
Journal volume & issue
Vol. 8, no. 1
pp. 1 – 14

Abstract

Read online

Behavioural tasks often require probability distributions to be inferred about task specific variables. Here, the authors demonstrate that generic neural networks can be trained using a simple error-based learning rule to perform such probabilistic computations efficiently without any need for task specific operations.