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..
Robust partially observable Markov decision process
Authors: Takayuki Osogami
ICML 2015 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Numerical experiments show that our point-based value iteration can adequately find robust policies. |
| Researcher Affiliation | Industry | Takayuki Osogami EMAIL IBM Research Tokyo, Tokyo, Japan |
| Pseudocode | Yes | Algorithm 1 Robust value iteration; Algorithm 2 Robust DP backup; Algorithm 3 Robust point-based DP backup |
| Open Source Code | No | The paper does not include any statement about releasing source code, nor does it provide a link to a code repository. |
| Open Datasets | No | The paper mentions using "Heaven and Hell, a standard instance of a POMDP" in its numerical experiments. However, it does not provide concrete access information (like a specific link, DOI, repository name, or formal citation with authors/year) for this or any other dataset or environment used. |
| Dataset Splits | No | The paper describes a reinforcement learning/planning problem setup and numerical experiments, but it does not specify any training, validation, or test dataset splits in terms of percentages, sample counts, or predefined external splits. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., GPU models, CPU types, memory) used for running the numerical experiments. |
| Software Dependencies | No | The paper does not specify any software dependencies with version numbers (e.g., Python, specific libraries, or solvers). |
| Experiment Setup | Yes | The agent moves one step at a time with the reward of 1 (unit cost). The agent obtains the reward of 1 upon reaching heaven or the reward of 10 upon reaching hell, and terminates the travel. The agent seeks to maximize the expected cumulative reward with the discount rate of γ = 0.9. When pe is large, the agent should directly go to an arbitrary ? , because the cost of going to ! for an observation pays off only when the observation is informative. A difficulty here is that pe is uncertain. |