Repository logo
Log in(current)
Repository logoMIT Open ScholarshipDSpace@MIT
  1. Home
  2. MIT Open Access Articles
  3. MIT Open Access Articles
  4. Evolutionary and Coevolutionary Multi-Agent Design Choices and Dynamics

Evolutionary and Coevolutionary Multi-Agent Design Choices and Dynamics

Thumbnail Image
Download
Name

3712255.3726712.pdf

Size

712.34 KB

Format

Adobe PDF

Checksum (MD5)

6ecebeb2508f3977890591c87073e930

Download all files submitted through automated deposit
3712255.3726712.zip (655.64 KB)
Author(s)
Hemberg, Erik
•
Moskal, Stephen
•
O'Reilly, Una-May
•
Liu
•
Fuller
Date Issued
August 11, 2025
Publisher
ACM|Genetic and Evolutionary Computation Conference
Citation
Erik Hemberg, Stephen Moskal, Una-May O'Reilly, Eric Liu, and Lucille Fuller. 2025. Evolutionary and Coevolutionary Multi-Agent Design Choices and Dynamics. In Proceedings of the Genetic and Evolutionary Computation Conference Companion (GECCO '25 Companion). Association for Computing Machinery, New York, NY, USA, 559–562.
Version
Final published version
Abstract
We investigate two representation alternatives for the controllers of teams of cyber agents. We combine these controller representations with different evolutionary algorithms, one of which introduces a novel LLM-supported mutation operator. Using a cyber security scenario, we evaluate agent learning when one side is trained to compete against a side that does not evolve and when two sides coevolve with each other. This allows us to quantify the relative merits and tradeoffs of representation and algorithm combinations in terms of team performance. The scenario also allows us to compare the performance impact and dynamics of coevolution versus evolution under different combinations. Across the algorithms and representations, we observe that coevolution reduces the performance highs and lows of both sides while it induces fluctuations on both sides. In contrast, when only one-side is optimized, performance peaks are higher and is more sustained than when both sides are optimized with coevolution.
Description
GECCO ’25 Companion, July 14–18, 2025, Malaga, Spain
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
https://hdl.handle.net/1721.1/162637
DOI of Published Version
https://doi.org/10.1145/3712255.3726712
Repository logo
PrivacyPermissionsAccessibilityContact us
Repository logo
Notify us about copyright concerns.