Constrained Deep Learning using Conditional Gradient and Applications in Computer Vision

03/17/2018
by   Sathya N. Ravi, et al.
0

A number of results have recently demonstrated the benefits of incorporating various constraints when training deep architectures in vision and machine learning. The advantages range from guarantees for statistical generalization to better accuracy to compression. But support for general constraints within widely used libraries remains scarce and their broader deployment within many applications that can benefit from them remains under-explored. Part of the reason is that Stochastic gradient descent (SGD), the workhorse for training deep neural networks, does not natively deal with constraints with global scope very well. In this paper, we revisit a classical first order scheme from numerical optimization, Conditional Gradients (CG), that has, thus far had limited applicability in training deep models. We show via rigorous analysis how various constraints can be naturally handled by modifications of this algorithm. We provide convergence guarantees and show a suite of immediate benefits that are possible -- from training ResNets with fewer layers but better accuracy simply by substituting in our version of CG to faster training of GANs with 50 better generalization guarantees using efficiently implementable forms of recently proposed regularizers.

READ FULL TEXT

page 9

page 12

research
08/18/2019

Towards Better Generalization: BP-SVRG in Training Deep Neural Networks

Stochastic variance-reduced gradient (SVRG) is a classical optimization ...
research
05/23/2019

Blockwise Adaptivity: Faster Training and Better Generalization in Deep Learning

Stochastic methods with coordinate-wise adaptive stepsize (such as RMSpr...
research
09/13/2017

Normalized Direction-preserving Adam

Optimization algorithms for training deep models not only affects the co...
research
01/16/2013

Revisiting Natural Gradient for Deep Networks

We evaluate natural gradient, an algorithm originally proposed in Amari ...
research
08/26/2018

Deep Learning: Computational Aspects

In this article we review computational aspects of Deep Learning (DL). D...
research
06/17/2020

Constraint-Based Regularization of Neural Networks

We propose a method for efficiently incorporating constraints into a sto...
research
07/05/2021

Generalization by design: Shortcuts to Generalization in Deep Learning

We take a geometrical viewpoint and present a unifying view on supervise...

Please sign up or login with your details

Forgot password? Click here to reset