Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..

Faster Fixed-Point Methods for Multichain MDPs

Authors: Matthew Zurek, Yudong Chen

NeurIPS 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental While this paper focuses on theoretical complexity analysis, we provide a few preliminary experimental examples.
Researcher Affiliation Academia Matthew Zurek Department of Computer Sciences University of Wisconsin Madison EMAIL Yudong Chen Department of Computer Sciences University of Wisconsin Madison EMAIL
Pseudocode Yes Algorithm 1 Approximately Shifted Halpern Iteration Algorithm 2 Halpern-Then-Picard Algorithm 3 Warm-Start Halpern-Then-Picard Algorithm 4 Approximately Shifted Halpern Iteration (General)
Open Source Code Yes Question: Does the paper provide open access to the data and code, with sufficient instructions to faithfully reproduce the main experimental results, as described in supplemental material? Answer: [Yes] Justification: All code used to produce the experimental results is provided.
Open Datasets No First we describe, for parameters k, T 1 and ε > 0, the parameterized MDP M(k, T) on which we run our experiments. M(k, T) has k + 1 states, where state 0 is absorbing (has only one action, which leads back to state 0).
Dataset Splits No The paper uses a custom-defined parameterized MDP M(k,T) for its experiments, which is described in detail, but it does not involve external datasets with explicit train/test/validation splits in the traditional sense.
Hardware Specification Yes All experiments were run on a single consumer laptop in less than a minute.
Software Dependencies No The paper does not specify any particular software libraries or their version numbers used for the experiments. It only states that the experiments were run on a "single consumer laptop".
Experiment Setup Yes First we describe, for parameters k, T 1 and ε > 0, the parameterized MDP M(k, T) on which we run our experiments. [...] We plot both ε = 1/2 and ε = 1/20, keeping T = 10, k = 300.