In this study, we present a mass-conserving variant of the LSTM and its application to arithmetic tasks, traffic forecasting, modeling a pendulum and rainfall-runoff modeling.
Abstract
The success of Convolutional Neural Networks (CNNs) in computer vision is mainly driven by their strong inductive bias, which is strong enough to allow CNNs to solve vision-related tasks with random weights, meaning without learning. Similarly, Long Short-Term Memory (LSTM) has a strong inductive bias towards storing information over time. However, many real-world systems are governed by conservation laws, which lead to the redistribution of particular quantities — e.g., in physical and economical systems. Our novel Mass-Conserving LSTM (MC-LSTM) adheres to these conservation laws by extending the inductive bias of LSTM to model the redistribution of those stored quantities. MC-LSTMs set a new state-of-the-art for neural arithmetic units at learning arithmetic operations, such as addition tasks,which have a strong conservation law, as the sum is constant over time. Further, MC-LSTM is applied to traffic forecasting, modeling a pendulum, and a large benchmark dataset in hydrology, where it sets a new state-of-the-art for predicting peak flows. In the hydrology example, we show that MC-LSTM states correlate with real world processes and are therefore interpretable.
Paper
Code
Citation
@inproceedings{hoedt2021mclstm,
title = {MC-LSTM: Mass-Conserving LSTM},
author = {Hoedt, Pieter-Jan and Kratzert, Frederik and Klotz, Daniel and Halmich, Christina and Holzleitner, Markus and Nearing, Grey S. and Hochreiter, Sepp and Klambauer, G\"{u}nter},
booktitle = {Proceedings of the 38th International Conference on Machine Learning},
pages = {4275--4286},
year = {2021},
editor = {Meila, Marina and Zhang, Tong},
volume = {139},
series = {Proceedings of Machine Learning Research},
month = {18--24 Jul},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v139/hoedt21a/hoedt21a.pdf},
url = {https://proceedings.mlr.press/v139/hoedt21a.html},
}