Towards Arbitrary Noise Augmentation - Deep Learning for Sampling from Arbitrary Probability Distributions
This provides a flexible alternative to existing sampling methods for noise modeling in novel sensors, though it is incremental as it builds on known neural network approaches.
The paper tackles the problem of sampling from arbitrary probability distributions for noise augmentation in deep learning by proposing a fully connected neural network that maps uniform samples to target distributions, minimizing Jensen-Shannon divergence. It demonstrates convergence and offers high sampling efficiency without requiring additional analytical or numerical calculations.
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.