Guiding Evolutionary AutoEncoder Training with Activation-Based Pruning Operators
Name
3712256.3726449.pdf
Size
3.43 MB
Format
Adobe PDF
Checksum (MD5)
311fd627cb34519603f980f2ecc62b14
Author(s)
Jorgensen, Steven
Hemberg, Erik
Toutouh, Jamal
O'Reilly, Una-May
Date Issued
July 13, 2025
Publisher
ACM|Genetic and Evolutionary Computation Conference
Citation
Steven Jorgensen, Erik Hemberg, Jamal Toutouh, and Una-May O'Reilly. 2025. Guiding Evolutionary AutoEncoder Training with Activation-Based Pruning Operators. In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO '25). Association for Computing Machinery, New York, NY, USA, 386–396.
Version
Final published version
Abstract
This study explores a novel approach to neural network pruning using evolutionary computation, focusing on simultaneously pruning the encoder and decoder of an autoencoder. We introduce two new mutation operators that use layer activations to guide weight pruning. Our findings reveal that one of these activation-informed operators outperforms random pruning, resulting in more efficient autoencoders with comparable performance to canonically trained models. Prior work has established that autoencoder training is effective and scalable with a spatial coevolutionary algorithm that cooperatively coevolves a population of encoders with a population of decoders, rather than one autoencoder. We evaluate how the same activity-guided mutation operators transfer to this context. We find that random pruning is better than guided pruning, in the coevolutionary setting. This suggests activation-based guidance proves more effective in low-dimensional pruning environments, where constrained sample spaces can lead to deviations from true uniformity in randomization. Conversely, population-driven strategies enhance robustness by expanding the total pruning dimensionality, achieving statistically uniform randomness that better preserves system dynamics. We experiment with pruning according to different schedules and present best combinations of operator and schedule for the canonical and coevolving populations cases.
Description
GECCO ’25, July 14–18, 2025, Malaga, Spain
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Terms of Use
Creative Commons Attribution
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/3712256.3726449
