# DeepMoD: Deep learning for Model Discovery in noisy data

## TL;DR

DeepMoD is a robust deep learning algorithm that discovers underlying partial differential equations from noisy spatio-temporal data using sparse regression, requiring minimal data and tolerating high noise levels.

## Contribution

It introduces a novel neural network approach that combines data approximation with sparse regression for PDE discovery without needing a training set.

## Key findings

- Works with as few as 100 samples
- Handles noise levels up to 75%
- Successfully applied to experimental data

## Abstract

We introduce DeepMoD, a Deep learning based Model Discovery algorithm. DeepMoD discovers the partial differential equation underlying a spatio-temporal data set using sparse regression on a library of possible functions and their derivatives. A neural network approximates the data and constructs the function library, but it also performs the sparse regression. This construction makes it extremely robust to noise, applicable to small data sets, and, contrary to other deep learning methods, does not require a training set. We benchmark our approach on several physical problems such as the Burgers', Korteweg-de Vries and Keller-Segel equations, and find that it requires as few as $\mathcal{O}(10^2)$ samples and works at noise levels up to $75\%$. Motivated by these results, we apply DeepMoD directly on noisy experimental time-series data from a gel electrophoresis experiment and find that it discovers the advection-diffusion equation describing this system.