Search Results for author: Geoffrey Roeder

Found 6 papers, 3 papers with code

Probabilistic Graphical Models and Tensor Networks: A Hybrid Framework

no code implementations29 Jun 2021 Jacob Miller, Geoffrey Roeder, Tai-Danae Bradley

We first prove that applying decoherence to the entirety of a BM model converts it into a discrete UGM, and conversely, that any subgraph of a discrete UGM can be represented as a decohered BM.

Tensor Networks

On Linear Identifiability of Learned Representations

no code implementations1 Jul 2020 Geoffrey Roeder, Luke Metz, Diederik P. Kingma

Identifiability is a desirable property of a statistical model: it implies that the true model parameters may be estimated to any desired precision, given sufficient computational resources and data.

Representation Learning

Efficient Amortised Bayesian Inference for Hierarchical and Nonlinear Dynamical Systems

1 code implementation28 May 2019 Geoffrey Roeder, Paul K. Grant, Andrew Phillips, Neil Dalchau, Edward Meeds

Our model class is a generalisation of nonlinear mixed-effects (NLME) dynamical systems, the statistical workhorse for many experimental sciences.

Bayesian Inference Zero-Shot Learning

Sticking the Landing: Simple, Lower-Variance Gradient Estimators for Variational Inference

1 code implementation NeurIPS 2017 Geoffrey Roeder, Yuhuai Wu, David Duvenaud

We propose a simple and general variant of the standard reparameterized gradient estimator for the variational evidence lower bound.

Variational Inference

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