Nov 20, 2023 · We study CVaR RL in low-rank MDPs with nonlinear function approximation. Low-rank MDPs assume the underlying transition kernel admits a low-rank ...
Feb 11, 2024 · This paper study risk-sensitive Reinforcement Learning under low-rank MDPs, where the transitions of MDPs admit a low-rank decomposition into two unknown low- ...
Nov 20, 2023 · Specifically, we present the first sample- efficient algorithm for optimizing the static CVaR metric that carefully balances the interplay.
We study risk-sensitive Reinforcement Learning (RL), where we aim to maximize the Conditional Value at Risk (CVaR) with a fixed risk tolerance τ.
This work designs a novel discretized Least-Squares Value Iteration algorithm for the CVaR objective as the planning oracle and shows that it can find the ...
Nov 20, 2023 · We study risk-sensitive Reinforcement Learning (RL), where we aim to maximize the Conditional Value at Risk (CVaR) with a fixed risk ...
Provably Efficient CVaR RL in Low-rank MDPs. Y. Zhao, W. Zhan, X. Hu, H. fung Leung, F. Farnia, W. Sun, and J. Lee. CoRR, (2023 ).
We study representation selection for a class of low-rank Markov Decision Processes (MDPs) where the transition kernel can be represented in a bilinear form.
We show partial coverage and realizability is enough for efficient model-based learning in offline RL; notable examples include low-rank MDPs, KNRs, and ...
Reinforcement learning (RL) under changing environment models many real-world applications via nonstationary Markov Decision Processes (MDPs), and hence.
Missing: CVaR | Show results with:CVaR