Path Sample-Analytic Gradient Estimators for Stochastic Binary Networks

06/04/2020
by   Alexander Shekhovtsov, et al.
0

In networks with binary activations and or binary weights the training by gradient descent is complicated as the model has piecewise constant response. We consider stochastic binary networks, obtained by adding noises in front of activations. The expected model response becomes a smooth function of parameters, its gradient is well defined but is challenging to estimate accurately. We propose a new method for this estimation problem combining sampling and analytic approximation steps. The method has a significantly reduced variance at the price of a small bias which gives a very practical tradeoff in comparison with existing unbiased and biased estimators. We further show that one extra linearization step leads to a deep straight-through estimator previously known only as an ad-hoc heuristic. We experimentally show higher accuracy in gradient estimation and demonstrate a more stable and better performing training in deep convolutional models with both proposed methods.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/31/2019

Perturbative estimation of stochastic gradients

In this paper we introduce a family of stochastic gradient estimation te...
research
06/11/2020

Reintroducing Straight-Through Estimators as Principled Methods for Stochastic Binary Networks

Training neural networks with binary weights and activations is a challe...
research
10/07/2021

Bias-Variance Tradeoffs in Single-Sample Binary Gradient Estimators

Discrete and especially binary random variables occur in many machine le...
research
05/15/2019

Hybrid Stochastic Gradient Descent Algorithms for Stochastic Nonconvex Optimization

We introduce a hybrid stochastic estimator to design stochastic gradient...
research
06/11/2014

Techniques for Learning Binary Stochastic Feedforward Neural Networks

Stochastic binary hidden units in a multi-layer perceptron (MLP) network...
research
02/28/2019

Deductive semiparametric estimation in Double-Sampling Designs with application to PEPFAR

Robust estimators in missing data problems often use semiparametric esti...
research
08/01/2019

Estimating the Standard Error of Cross-Validation-Based Estimators of Classification Rules Performance

First, we analyze the variance of the Cross Validation (CV)-based estima...

Please sign up or login with your details

Forgot password? Click here to reset