In this section
@inproceedings{Lim:2022:10.1145/3520304.3528927,
author = {Lim, B and Allard, M and Grillotti, L and Cully, A},
doi = {10.1145/3520304.3528927},
pages = {128--131},
publisher = {Association for Computing Machinery},
title = {QDax: on the benefits of massive parallelization for quality-diversity},
url = {http://dx.doi.org/10.1145/3520304.3528927},
year = {2022}
}
TY - CPAPER
AB - Quality-Diversity (QD) algorithms are a well-known approach to generate large collections of diverse and high-quality policies. However, QD algorithms are also known to be data-inefficient, requiring large amounts of computational resources and are slow when used in practice for robotics tasks. Policy evaluations are already commonly performed in parallel to speed up QD algorithms but have limited capabilities on a single machine as most physics simulators run on CPUs. With recent advances in simulators that run on accelerators, thousands of evaluations can be performed in parallel on single GPU/TPU. In this paper, we present QDax, an implementation of MAP-Elites which leverages massive parallelism on accelerators to make QD algorithms more accessible. We show that QD algorithms are ideal candidates and can scale with massive parallelism to be run at interactive timescales. The increase in parallelism does not significantly affect the performance of QD algorithms, while reducing experiment runtimes by two factors of magnitudes, turning days of computation into minutes. These results show that QD can now benefit from hardware acceleration, which contributed significantly to the bloom of deep learning.
AU - Lim,B
AU - Allard,M
AU - Grillotti,L
AU - Cully,A
DO - 10.1145/3520304.3528927
EP - 131
PB - Association for Computing Machinery
PY - 2022///
SP - 128
TI - QDax: on the benefits of massive parallelization for quality-diversity
UR - http://dx.doi.org/10.1145/3520304.3528927
UR - https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:001035469400052&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=a2bf6146997ec60c407a63945d4e92bb
UR - http://hdl.handle.net/10044/1/115720
ER -
