A discounted value-iteration algorithm with visited-state clipping and deviation-controlled updates achieves ~O(sp(v*) sqrt(d^3 T)) regret for infinite-horizon average-reward linear MDPs with computational cost independent of the state-space size.
Inven tory management in supply chains: a re- inforcement learning approach
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Computationally Efficient Algorithm for Infinite-Horizon Average-Reward Linear MDPs
A discounted value-iteration algorithm with visited-state clipping and deviation-controlled updates achieves ~O(sp(v*) sqrt(d^3 T)) regret for infinite-horizon average-reward linear MDPs with computational cost independent of the state-space size.