Poster

Asymmetric Heavy Tails and Implicit Bias in Gaussian Noise Injections

Alexander D Camuto · Xiaoyu Wang · Lingjiong Zhu · Christopher Holmes · Mert Gurbuzbalaban · Umut Simsekli

Keywords: [ Deep Learning Theory ]

[ Abstract ]
[ Paper ]
[ Visit Poster at Spot A6 in Virtual World ] [ Visit Poster at Spot D1 in Virtual World ]
Wed 21 Jul 9 p.m. PDT — 11 p.m. PDT
 
Spotlight presentation: Deep Learning Theory 4
Wed 21 Jul 5 p.m. PDT — 6 p.m. PDT

Abstract:

Gaussian noise injections (GNIs) are a family of simple and widely-used regularisation methods for training neural networks, where one injects additive or multiplicative Gaussian noise to the network activations at every iteration of the optimisation algorithm, which is typically chosen as stochastic gradient descent (SGD). In this paper, we focus on the so-called implicit effect' of GNIs, which is the effect of the injected noise on the dynamics of SGD. We show that this effect induces an \emph{asymmetric heavy-tailed noise} on SGD gradient updates. In order to model this modified dynamics, we first develop a Langevin-like stochastic differential equation that is driven by a general family of \emph{asymmetric} heavy-tailed noise. Using this model we then formally prove that GNIs induce animplicit bias', which varies depending on the heaviness of the tails and the level of asymmetry. Our empirical results confirm that different types of neural networks trained with GNIs are well-modelled by the proposed dynamics and that the implicit effect of these injections induces a bias that degrades the performance of networks.

Chat is not available.