paper-with-me

Papers

Contrastive Divergence Learning is a Time Reversal Adversarial Game

2020-12-06 · ICLR 2021 1 · Omer Yair, Tomer Michaeli

Contrastive divergence (CD) learning is a classical method for fitting unnormalized statistical models to data samples. Despite its wide-spread use, the convergence properties of this algorithm are still not well understood. The main source of difficulty is an unjustified approximation which has been used to derive the gradient of the loss. In this paper, we present an alternative derivation of CD that does not require any approximation and sheds new light on the objective that is actually being optimized by the algorithm. Specifically, we show that CD is an adversarial learning procedure, where a discriminator attempts to classify whether a Markov chain generated from the model has been time-reversed. Thus, although predating generative adversarial networks (GANs) by more than a decade, CD is, in fact, closely related to these techniques. Our derivation settles well with previous observations, which have concluded that CD's update steps cannot be expressed as the gradients of any fixed objective function. In addition, as a byproduct, our derivation reveals a simple correction that can be used as an alternative to Metropolis-Hastings rejection, which is required when the underlying Markov chain is inexact (e.g. when using Langevin dynamics with a large step).

📄 PDF Abstract BibTeX arXiv:2012.03295

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Adversarial Training using Contrastive Divergence

2021-01-01 · Hongjun Wang, Guanbin Li, Liang Lin

To protect the security of machine learning models against adversarial examples, adversarial training becomes the most popular and powerful strategy against various adversarial attacks by injecting adversarial examples i…

f-Domain-Adversarial Learning: Theory and Algorithms for Unsupervised Domain Adaptation with Neural Networks

2021-01-01 · David Acuna, Guojun Zhang, Marc T Law, Sanja Fidler

The problem of unsupervised domain adaptation arises in a variety of practical applications where the distribution of the training samples differs from those used at test time. The existing theory of domain adaptation de…

Domain AdaptationGeneralization BoundsLearning TheoryUnsupervised Domain Adaptation

Training Deep Energy-Based Models with f-Divergence Minimization

2020-03-06 · ICML 2020 1 · Lantao Yu, Yang song, Jiaming Song, Stefano Ermon

Deep energy-based models (EBMs) are very flexible in distribution parametrization but computationally challenging because of the intractable partition function. They are typically trained via maximum likelihood, using co…

Test-time Detection and Repair of Adversarial Samples via Masked Autoencoder

2023-03-22 · Yun-Yun Tsai, Ju-Chin Chao, Albert Wen, Zhaoyuan Yang 외

Training-time defenses, known as adversarial training, incur high training costs and do not generalize to unseen attacks. Test-time defenses solve these issues but most existing test-time defenses require adapting the mo…

Contrastive LearningManagement

REDAT: Accent-Invariant Representation for End-to-End ASR by Domain Adversarial Training with Relabeling

2020-12-14 · Hu Hu, Xuesong Yang, Zeynab Raeesy, Jinxi Guo 외

Accents mismatching is a critical problem for end-to-end ASR. This paper aims to address this problem by building an accent-robust RNN-T system with domain adversarial training (DAT). We unveil the magic behind DAT and p…

Clustering