Solving Uncertain MDPs by Reusing State Information and Plans

Authors: Ping Hou, William Yeoh, Tran Cao Son

AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We experimentally show that our approach is typically faster than replanning from scratch and we also provide a way to estimate the amount of speedup based on the amount of information being reused.
Researcher Affiliation Academia Ping Hou Department of Computer Science New Mexico State University Las Cruces, NM 88003 phou@cs.nmsu.edu William Yeoh Department of Computer Science New Mexico State University Las Cruces, NM 88003 wyeoh@cs.nmsu.edu Tran Cao Son Department of Computer Science New Mexico State University Las Cruces, NM 88003 tson@cs.nmsu.edu
Pseudocode Yes Algorithm 1: INCREMENTAL-MDP(M) and Algorithm 2: FIND-REUSABLE-STATES(I, S , Y )
Open Source Code No The paper does not provide any statement or link regarding the public availability of its source code.
Open Datasets Yes We run the algorithms on two sets of domains: (i) a multi-layer navigation domain similar to that in the literature (Dai et al. 2011), and (ii) the domains in the ICAPS 2011 International Probabilistic Planning Competition (IPPC).4
Dataset Splits No The paper describes experimental setups and problem generation for the domains used but does not specify explicit training, validation, or test dataset splits or percentages.
Hardware Specification Yes We conducted our experiments on a dual-core 2.53 GHz machine with 4GB of RAM.
Software Dependencies No The paper mentions the MDP algorithms used (e.g., VI, TVI, UCT) but does not specify any software dependencies or their version numbers (e.g., Python, PyTorch, specific solver versions).
Experiment Setup Yes We set X to 50, Y to 500, and R to 50.