147 citations · 196 across the 11 of their papers we have counts for
7 papers · 1 filter
An online sequence-to-sequence model for noisy speech recognition
Chung-Cheng Chiu, Dieterich Lawson, Yuping Luo +4
Generative models have long been the dominant approach for speech recognition. The success of these models however relies on the use of sophisticated recipes and complicated machin…
Filtering Variational Objectives
Chris J. Maddison, Dieterich Lawson, George Tucker +5
When used as a surrogate objective for maximum likelihood estimation in latent variable models, the evidence lower bound (ELBO) produces state-of-the-art results. Inspired by this,…
Learning Hard Alignments with Variational Inference
Dieterich Lawson, Chung-Cheng Chiu, George Tucker +3
There has recently been significant interest in hard attention models for tasks such as object recognition, visual captioning and speech recognition. Hard attention can offer benef…
Particle Value Functions
Chris J. Maddison, Dieterich Lawson, George Tucker +4
The policy gradients of the expected return objective can react slowly to rare rewards. Yet, in some cases agents may wish to emphasize the low or high returns regardless of their…
REBAR: Low-variance, unbiased gradient estimates for discrete latent variable models
George Tucker, Andriy Mnih, Chris J. Maddison +2
Learning in models with discrete latent variables is challenging due to high variance gradient estimators. Generally, approaches have relied on control variates to reduce the varia…
Changing Model Behavior at Test-Time Using Reinforcement Learning
Augustus Odena, Dieterich Lawson, Christopher Olah
Machine learning models are often used at test-time subject to constraints and trade-offs not present at training-time. For example, a computer vision model operating on an embedde…
Training a Subsampling Mechanism in Expectation
Colin Raffel, Dieterich Lawson
We describe a mechanism for subsampling sequences and show how to compute its expected output so that it can be trained with standard backpropagation. We test this approach on a si…