Towards Arbitrary Noise Augmentation - Deep Learning for Sampling from Arbitrary Probability Distributions

TL;DR

This paper introduces a neural network-based method for sampling from arbitrary probability distributions, enabling accurate noise modeling in deep learning reconstruction without complex calculations.

Contribution

It proposes a fully connected neural network that learns to generate samples from any known distribution by minimizing Jensen-Shannon divergence, offering an efficient alternative to traditional sampling methods.

Findings

Model converges to target distribution during training

Provides high sampling efficiency and flexibility

Outperforms traditional sampling methods in simplicity and applicability

Abstract

Accurate noise modelling is important for training of deep learning reconstruction algorithms. While noise models are well known for traditional imaging techniques, the noise distribution of a novel sensor may be difficult to determine a priori. Therefore, we propose learning arbitrary noise distributions. To do so, this paper proposes a fully connected neural network model to map samples from a uniform distribution to samples of any explicitly known probability density function. During the training, the Jensen-Shannon divergence between the distribution of the model's output and the target distribution is minimized. We experimentally demonstrate that our model converges towards the desired state. It provides an alternative to existing sampling methods such as inversion sampling, rejection sampling, Gaussian mixture models and Markov-Chain-Monte-Carlo. Our model has high sampling efficiency and is easily applied to any probability distribution, without the need of further analytical or numerical calculations.