# TATi-Thermodynamic Analytics ToolkIt: TensorFlow-based software for posterior sampling in machine learning applications

## TL;DR

This paper introduces TATi, a TensorFlow-based software toolkit that enables efficient Bayesian posterior sampling for neural networks using advanced MCMC methods, demonstrated on MNIST data.

## Contribution

The paper presents a new software toolkit, TATi, that facilitates posterior sampling in neural networks with novel preconditioning techniques and visualization capabilities.

## Key findings

- Sampling efficiency improved with ensemble quasi-Newton preconditioning.
- Visualization of neural network loss landscape on MNIST.
- Demonstration of Bayesian parametrization in neural networks.

## Abstract

With the advent of GPU-assisted hardware and maturing high-efficiency software platforms such as TensorFlow and PyTorch, Bayesian posterior sampling for neural networks becomes plausible. In this article we discuss Bayesian parametrization in machine learning based on Markov Chain Monte Carlo methods, specifically discretized stochastic differential equations such as Langevin dynamics and extended system methods in which an ensemble of walkers is employed to enhance sampling. We provide a glimpse of the potential of the sampling-intensive approach by studying (and visualizing) the loss landscape of a neural network applied to the MNIST data set. Moreover, we investigate how the sampling efficiency itself can be significantly enhanced through an ensemble quasi-Newton preconditioning method. This article accompanies the release of a new TensorFlow software package, the Thermodynamic Analytics ToolkIt, which is used in the computational experiments.