Distributional Multi-Objective Decision Making
Authors: Willem Röpke, Conor F. Hayes, Patrick Mannion, Enda Howley, Ann Nowé, Diederik M. Roijers
IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Through experiments, we demonstrate the feasibility and effectiveness of these methods, making this a valuable new approach for decision support in real-world problems. |
| Researcher Affiliation | Collaboration | 1Vrije Universiteit Brussel, Brussels, Belgium 2University of Galway, Galway, Ireland 3City of Amsterdam, Amsterdam, The Netherlands |
| Pseudocode | Yes | Algorithm 1 CDPrune Algorithm 2 DIMOQ |
| Open Source Code | Yes | All code is available at https://github.com/wilrop/ distributional-dominance. |
| Open Datasets | No | The paper states, 'We evaluate DIMOQ (Algorithm 2) and CDPrune (Algorithm 1) on randomly generated MOMDPs of different sizes shown in Table 1.' It generates its own data and does not provide access to a public or open dataset of these generated MOMDPs. |
| Dataset Splits | No | The paper uses 'randomly generated MOMDPs' and describes their configuration in Table 1 but does not specify any train/validation/test splits. |
| Hardware Specification | Yes | All experiments were run on a single core of an Intel Xeon Gold 6148 processor, with a maximum RAM requirement of 2GB. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, specific libraries and their versions). |
| Experiment Setup | Yes | For each size category, we repeat the experiment with seeds one through five and perform 50, 000 random walks to estimate T followed by 2, 000 training episodes. All experiments considered two objectives, used a discount factor of 1 and limited the precision of distributions to three decimals. |