IBAL framework constructs information-theoretic adversarial attacks on agent observations and actions to train MARL agents that remain robust to interaction disruptions and agent-missing scenarios.
Reward poisoning in reinforcement learning: Attacks against un- known learners in unknown environments.arXiv preprint arXiv:2102.08492
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
fields
cs.LG 4verdicts
UNVERDICTED 4representative citing papers
Preference poisoning against log-linear DPO reduces to a binary sparse approximation problem solved by lattice-reduction (BAL-A) and matching-pursuit (BMP-A) algorithms that carry recovery guarantees.
Reward poisoning in linear MDPs is attackable if and only if a precise structural condition holds, drawing a sharp line between vulnerable and intrinsically robust instances.
Wolfpack attack framework disrupts MARL cooperation by targeting initial and assisting agents; WALL trains robust policies against it with reported experimental gains.
citing papers explorer
-
Interaction-Breaking Adversarial Learning Framework for Robust Multi-Agent Reinforcement Learning
IBAL framework constructs information-theoretic adversarial attacks on agent observations and actions to train MARL agents that remain robust to interaction disruptions and agent-missing scenarios.
-
Efficient Preference Poisoning Attack on Offline RLHF
Preference poisoning against log-linear DPO reduces to a binary sparse approximation problem solved by lattice-reduction (BAL-A) and matching-pursuit (BMP-A) algorithms that carry recovery guarantees.
-
When Can You Poison Rewards? A Tight Characterization of Reward Poisoning in Linear MDPs
Reward poisoning in linear MDPs is attackable if and only if a precise structural condition holds, drawing a sharp line between vulnerable and intrinsically robust instances.
-
Wolfpack Adversarial Attack for Robust Multi-Agent Reinforcement Learning
Wolfpack attack framework disrupts MARL cooperation by targeting initial and assisting agents; WALL trains robust policies against it with reported experimental gains.