TyXe: Pyro-based Bayesian neural nets for Pytorch

TL;DR

TyXe is a flexible, Pyro-based library for Bayesian neural networks in Pytorch, enabling easy experimentation with priors, inference, and architectures for uncertainty estimation across various models.

Contribution

TyXe introduces a modular, architecture-agnostic framework for Bayesian neural networks in Pytorch, simplifying the integration of Bayesian methods into existing models.

Findings

Demonstrates ease of converting deterministic models to Bayesian versions

Supports a wide range of models including CNNs, GNNs, and NeRFs

Provides practical tools for uncertainty estimation in deep learning

Abstract

We introduce TyXe, a Bayesian neural network library built on top of Pytorch and Pyro. Our leading design principle is to cleanly separate architecture, prior, inference and likelihood specification, allowing for a flexible workflow where users can quickly iterate over combinations of these components. In contrast to existing packages TyXe does not implement any layer classes, and instead relies on architectures defined in generic Pytorch code. TyXe then provides modular choices for canonical priors, variational guides, inference techniques, and layer selections for a Bayesian treatment of the specified architecture. Sampling tricks for variance reduction, such as local reparameterization or flipout, are implemented as effect handlers, which can be applied independently of other specifications. We showcase the ease of use of TyXe to explore Bayesian versions of popular models from various libraries: toy regression with a pure Pytorch neural network; large-scale image classification with torchvision ResNets; graph neural networks based on DGL; and Neural Radiance Fields built on top of Pytorch3D. Finally, we provide convenient abstractions for variational continual learning. In all cases the change from a deterministic to a Bayesian neural network comes with minimal modifications to existing code, offering a broad range of researchers and practitioners alike practical access to uncertainty estimation techniques. The library is available at https://github.com/TyXe-BDL/TyXe.