Meta-Learned Adaptive Memory Filtering for Robust Multi-Agent Collaboration
Frontiers in Interdisciplinary Applied Science · 15.03.2026 · 3 (1) · 129-137
DOI: 10.71465/fias758 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Interdisciplinary Applied Science, online erschienen am 15.03.2026. Verfasst von Ethan Campbell und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Static defense rules may fail under evolving poisoning strategies. This work introduces a meta-learning framework that adapts memory filtering parameters based on historical contamination patterns. A meta-optimizer updates filtering thresholds using second-order gradient estimation across poisoning episodes. The framework was evaluated on cooperative resource management simulations with 50 agents across 200 poisoning scenarios. Adaptive filtering reduced cumulative task performance loss by 44.8% compared with fixed-threshold filtering. Convergence speed improved by 19.5% under dynamic attack conditions. Meta-adaptive filtering enhances resilience against non-stationary memory poisoning strategies.
Zitieren
APA
Campbell, E., Martin, S., & Parker, N. (2026). Meta-Learned Adaptive Memory Filtering for Robust Multi-Agent Collaboration. Frontiers in Interdisciplinary Applied Science, 3(1), 129-137. https://doi.org/10.71465/fias758
Vancouver
Campbell E, Martin S, Parker N. Meta-Learned Adaptive Memory Filtering for Robust Multi-Agent Collaboration. Frontiers in Interdisciplinary Applied Science. 2026;3(1):129-137. doi:10.71465/fias758
BibTeX
@article{Campbell2026,
author = {Campbell, Ethan and Martin, Sophie and Parker, Noah},
title = {Meta-Learned Adaptive Memory Filtering for Robust Multi-Agent Collaboration},
journal = {Frontiers in Interdisciplinary Applied Science},
year = {2026},
volume = {3},
number = {1},
pages = {129--137},
doi = {10.71465/fias758}
}
EUVIMED