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 [1].
On Markov Games Played by Bayesian and Boundedly-Rational Players
Authors: Muthukumaran Chandrasekaran, Yingke Chen, Prashant Doshi
AAAI 2017 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The computation of finding such equilibria is formalized as a constraint satisfaction problem and its effectiveness is demonstrated on two cooperative domains. We implemented the MAC3 algorithm for obtaining MPFLE as discussed earlier. We experiment with two benchmark problem domains: n-agent multiple access broadcast channel (n MABC) (Hansen, Bernstein, and Zilberstein 2004) (|N| = 2 to 5; H = 1 to 5; |S| = 4; |A| = 4; |Xi N| up to 4; i N |Θi| up to 1024) and sequential level-based foraging... |
| Researcher Affiliation | Academia | Muthukumaran Chandrasekaran THINC Lab, University of Georgia Athens, GA, USA EMAIL Yingke Chen Sichuan University Chengdu, China EMAIL Prashant Doshi THINC Lab, University of Georgia Athens, GA, USA EMAIL |
| Pseudocode | No | No explicitly labeled pseudocode or algorithm block was found. |
| Open Source Code | No | No clear statement or link providing open-source code for the methodology described in the paper was found. |
| Open Datasets | Yes | We experiment with two benchmark problem domains: n-agent multiple access broadcast channel (n MABC) (Hansen, Bernstein, and Zilberstein 2004) ... and sequential level-based foraging... (Albrecht and Ramamoorthy 2013) |
| Dataset Splits | No | No specific details on training/test/validation dataset splits (e.g., percentages, sample counts, or explicit cross-validation setup for their own experiments) were provided. The term 'Validation' is used in a general sense to describe checking their approach, not a dataset split. |
| Hardware Specification | Yes | Our computing configuration included an Intel Xeon 2.67GHz processor, 12 GB RAM and Linux. |
| Software Dependencies | No | The paper mentions using 'MAC3' as an algorithm ('We used the generic procedure described in an efficient arc consistency algorithm called MAC3 (Liu 1998) to solve the CSP.'), but no specific version number for MAC3 or any other software dependencies are provided. |
| Experiment Setup | Yes | We also found 12 and 17 ϵ-MPFLE for ϵ = 0.17 and 0.33 respectively. Figure 3: Impact of parameters on performance. Time taken to compute: (top) MPFLE in 2MABC and 3MABC for varying τ and ϵ at H = 3... The time taken to compute the first ϵ-MPFLE for varying horizons and two coarse discretizations is shown. Run time decreases by about two orders of magnitude as the discretization gets coarser for H = 2. |