Training Binary Neural Networks using the Bayesian Learning Rule

Code

• github.com/team-approx-bayes/BayesBiNN
  OfficialIn paperpytorch★ 41
• github.com/intellhave/adaste
  pytorch★ 9
• github.com/prateekstark/training-binary-neural-network
  pytorch★ 3
• github.com/mengxiangming/BayesBiNN
  pytorch★ 0

Abstract

Neural networks with binary weights are computation-efficient and hardware-friendly, but their training is challenging because it involves a discrete optimization problem. Surprisingly, ignoring the discrete nature of the problem and using gradient-based methods, such as the Straight-Through Estimator, still works well in practice. This raises the question: are there principled approaches which justify such methods? In this paper, we propose such an approach using the Bayesian learning rule. The rule, when applied to estimate a Bernoulli distribution over the binary weights, results in an algorithm which justifies some of the algorithmic choices made by the previous approaches. The algorithm not only obtains state-of-the-art performance, but also enables uncertainty estimation for continual learning to avoid catastrophic forgetting. Our work provides a principled approach for training binary neural networks which justifies and extends existing approaches.

Tasks

Reproductions