
Lossy Compression for Lossless Prediction
Most data is automatically collected and only ever "seen" by algorithms....
read it

Learning to Extend Program Graphs to WorkinProgress Code
Source code spends most of its time in a broken or incomplete state duri...
read it

Improving Lossless Compression Rates via Monte Carlo BitsBack Coding
Latent variable models have been successfully applied in lossless compre...
read it

Oops I Took A Gradient: Scalable Sampling for Discrete Distributions
We propose a general and scalable approximate sampling strategy for prob...
read it

RaoBlackwellizing the StraightThrough GumbelSoftmax Gradient Estimator
Gradient estimation in models with discrete latent variables is a challe...
read it

Learning Branching Heuristics for Propositional Model Counting
Propositional model counting or #SAT is the problem of computing the num...
read it

Gradient Estimation with Stochastic Softmax Tricks
The GumbelMax trick is the basis of many relaxed gradient estimators. T...
read it

On Empirical Comparisons of Optimizers for Deep Learning
Selecting an optimizer is a central step in the contemporary deep learni...
read it

Direct Policy Gradients: Direct Optimization of Policies in Discrete Action Spaces
Direct optimization is an appealing approach to differentiating through ...
read it

Hierarchical Representations with Poincaré Variational AutoEncoders
The Variational AutoEncoder (VAE) model has become widely popular as a ...
read it

Doubly Reparameterized Gradient Estimators for Monte Carlo Objectives
Deep latent variable models have become a popular model choice due to th...
read it

Hamiltonian Descent Methods
We propose a family of optimization methods that achieve linear converge...
read it

Conditional Neural Processes
Deep neural networks excel at function approximation, yet they are typic...
read it

Tighter Variational Bounds are Not Necessarily Better
We provide theoretical and empirical evidence that using tighter evidenc...
read it

Filtering Variational Objectives
When used as a surrogate objective for maximum likelihood estimation in ...
read it

REBAR: Lowvariance, unbiased gradient estimates for discrete latent variable models
Learning in models with discrete latent variables is challenging due to ...
read it

The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
The reparameterization trick enables optimizing large scale stochastic c...
read it

Move Evaluation in Go Using Deep Convolutional Neural Networks
The game of Go is more challenging than other board games, due to the di...
read it

A* Sampling
The problem of drawing samples from a discrete distribution can be conve...
read it

Structured Generative Models of Natural Source Code
We study the problem of building generative models of natural source cod...
read it
Chris J. Maddison
is this you? claim profile