Search Results for author: Charlie Nash

Found 11 papers, 6 papers with code

Variable-rate discrete representation learning

no code implementations10 Mar 2021 Sander Dieleman, Charlie Nash, Jesse Engel, Karen Simonyan

Semantically meaningful information content in perceptual signals is usually unevenly distributed.

Representation Learning

Generating Images with Sparse Representations

1 code implementation5 Mar 2021 Charlie Nash, Jacob Menick, Sander Dieleman, Peter W. Battaglia

The high dimensionality of images presents architecture and sampling-efficiency challenges for likelihood-based generative models.

Colorization Image Colorization +2

PolyGen: An Autoregressive Generative Model of 3D Meshes

1 code implementation ICML 2020 Charlie Nash, Yaroslav Ganin, S. M. Ali Eslami, Peter W. Battaglia

Polygon meshes are an efficient representation of 3D geometry, and are of central importance in computer graphics, robotics and games development.

3D Shape Generation Surface Reconstruction

Autoregressive Energy Machines

1 code implementation11 Apr 2019 Charlie Nash, Conor Durkan

We propose the Autoregressive Energy Machine, an energy-based model which simultaneously learns an unnormalized density and computes an importance-sampling estimate of the normalizing constant for each conditional in an autoregressive decomposition.

Density Estimation valid

Inverting Supervised Representations with Autoregressive Neural Density Models

no code implementations1 Jun 2018 Charlie Nash, Nate Kushman, Christopher K. I. Williams

In addition, we can use these inversion models to estimate the mutual information between a model's inputs and its intermediate representations, thus quantifying the amount of information preserved by the network at different stages.

Density Estimation

Autoencoders and Probabilistic Inference with Missing Data: An Exact Solution for The Factor Analysis Case

1 code implementation11 Jan 2018 Christopher K. I. Williams, Charlie Nash, Alfredo Nazábal

We show how to calculate exactly the latent posterior distribution for the factor analysis (FA) model in the presence of missing data, and note that this solution implies that a different encoder network is required for each pattern of missingness.


Generative Entity Networks: Disentangling Entitites and Attributes in Visual Scenes using Partial Natural Language Descriptions

no code implementations ICLR 2018 Charlie Nash, Sebastian Nowozin, Nate Kushman

Using the Shapeworld dataset, we show that our representation both enables a better generative model of images, leading to higher quality image samples, as well as creating more semantically useful representations that improve performance over purely dicriminative models on a simple natural language yes/no question answering task.

Question Answering

Cannot find the paper you are looking for? You can Submit a new open access paper.