• Classified by Topic • Classified by Publication Type • Sorted by Date • Sorted by First Author Last Name • Classified by Funding Source •
Dynamic Sparse Training for Deep Reinforcement Learning.
Ghada Sokar, Elena
Mocanu, Decebal Constantin Mocanu, Mykola Pechenizkiy,
and Peter Stone.
In Proceedings of the 31st International Joint Conference
on Artificial Intelligence, July 2022.
arXiv version with the appendix
[PDF]3.7MB [slides.pptx]16.7MB
Deep reinforcement learning (DRL) agents are trained through trial-and-error interactions with the environment. This leads to a long training time for dense neural networks to achieve good performance. Hence, prohibitive computation and memory resources are consumed. Recently, learning efficient DRL agents has received increasing attention. Yet, current methods focus on accelerating inference time. In this paper, we introduce for the first time a dynamic sparse training approach for deep reinforcement learning to accelerate the training process. The proposed approach trains a sparse neural network from scratch and dynamically adapts its topology to the changing data distribution during training. Experiments on continuous control tasks show that our dynamic sparse agents achieve higher performance than the equivalent dense methods, reduce the parameter count and floating-point operations (FLOPs) by 50%, and have a faster learning speed that enables reaching the performance of dense agents with 40-50% reduction in the training steps.
@InProceedings{IJCAI22,
author="Ghada Sokar and Elena Mocanu and Decebal Constantin Mocanu and Mykola Pechenizkiy and Peter Stone",
title="Dynamic Sparse Training for Deep Reinforcement Learning",
booktitle="Proceedings of the 31st International Joint Conference on Artificial Intelligence",
location="Vienna, Austria",
month="July",
year="2022",
abstract={
Deep reinforcement learning (DRL) agents are trained
through trial-and-error interactions with the
environment. This leads to a long training time for dense
neural networks to achieve good performance. Hence,
prohibitive computation and memory resources are
consumed. Recently, learning efficient DRL agents has
received increasing attention. Yet, current methods focus
on accelerating inference time. In this paper, we
introduce for the first time a dynamic sparse training
approach for deep reinforcement learning to accelerate the
training process. The proposed approach trains a sparse
neural network from scratch and dynamically adapts its
topology to the changing data distribution during
training. Experiments on continuous control tasks show
that our dynamic sparse agents achieve higher performance
than the equivalent dense methods, reduce the parameter
count and floating-point operations (FLOPs) by 50%, and
have a faster learning speed that enables reaching the
performance of dense agents with 40-50% reduction in the
training steps.
},
wwwnote={<a href="https://arxiv.org/pdf/2106.04217.pdf">arXiv version with the appendix</a>},
}
Generated by bib2html.pl (written by Patrick Riley ) on Thu Oct 23, 2025 16:14:16