Sara Klein

2 accepted papers

2025

Structure Matters: Dynamic Policy Gradient

Sara Klein, Xiangyuan Zhang, Tamer Basar, Simon Weissmann, Leif Döring

NeurIPS 2025poster

In this work, we study $\gamma$-discounted infinite-horizon tabular Markov decision processes (MDPs) and introduce a framework called dynamic policy gradient (DynPG). The framework directly integrates dynamic programming with (any) policy gradient method, explicitly leveraging the Markovian property…

Cited by 0 Source ↗Scholar ↗

2024

Beyond Stationarity: Convergence Analysis of Stochastic Softmax Policy Gradient Methods

Sara Klein, Simon Weissmann, Leif Döring

ICLR 2024poster

Markov Decision Processes (MDPs) are a formal framework for modeling and solving sequential decision-making problems. In finite time horizons such problems are relevant for instance for optimal stopping or specific supply chain problems, but also in the training of large language models. In contrast…

Cited by 8 Source ↗PDF ↗Scholar ↗