CA-MIQ combines a novelty-driven intrinsic critic with a priority-shift detector and selective value resets, achieving about four times higher mission success than baseline Q-learning after priority changes in a simulated SAR gridworld.
Regret bounds for information-directed reinforcement learning,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Learning What Matters Now: A Dual-Critic Context-Aware RL Framework for Priority-Driven Information Gain
CA-MIQ combines a novelty-driven intrinsic critic with a priority-shift detector and selective value resets, achieving about four times higher mission success than baseline Q-learning after priority changes in a simulated SAR gridworld.