COMP-PHLGSep 30, 2020

TensorBNN: Bayesian Inference for Neural Networks using Tensorflow

arXiv:2009.14393v31 citations
Originality Synthesis-oriented
AI Analysis

This provides a tool for researchers and practitioners needing probabilistic uncertainty estimates in neural networks, but it is incremental as it builds on existing Bayesian methods with TensorFlow integration.

The authors tackled the problem of performing Bayesian inference for neural networks by developing TensorBNN, a TensorFlow-based package that uses Hamiltonian Monte Carlo to sample the posterior density, enabling efficient GPU-accelerated training and prediction.

TensorBNN is a new package based on TensorFlow that implements Bayesian inference for modern neural network models. The posterior density of neural network model parameters is represented as a point cloud sampled using Hamiltonian Monte Carlo. The TensorBNN package leverages TensorFlow's architecture and training features as well as its ability to use modern graphics processing units (GPU) in both the training and prediction stages.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes