Point-Based POMDP Solving with Factored Value Function Approximation
Authors: Tiago Veiga, Matthijs Spaan, Pedro Lima
AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We experimentally verify our contributions and show that they have the potential to improve point-based methods in policy quality and solution size. ... To test the viability of the ideas presented in this paper, we performed a series of experiments. ... Our experiments have been performed with a Matlab implementation of Perseus (Spaan and Vlassis 2005) using basis functions to approximate value functions. |
| Researcher Affiliation | Academia | Tiago S. Veiga Institute for Systems and Robotics Instituto Superior T ecnico Universidade de Lisboa Lisbon, Portugal tsveiga@isr.ist.utl.pt Matthijs T. J. Spaan Delft University of Technology Delft, The Netherlands m.t.j.spaan@tudelft.nl Pedro U. Lima Institute for Systems and Robotics Instituto Superior T ecnico Universidade de Lisboa Lisbon, Portugal pal@isr.ist.utl.pt |
| Pseudocode | Yes | Algorithm 1 Point-based backup with linear value function approximation |
| Open Source Code | No | The paper does not provide any statement or link indicating that the source code for the described methodology is open-source or publicly available. |
| Open Datasets | Yes | The benchmark problem domains for our tests are a variation of the fire fighting problem (Oliehoek et al. 2008) and the network management problem (Poupart 2005). |
| Dataset Splits | No | The paper mentions using 'a belief set of 500 belief points' and 'results are averaged over 10 times the number of start positions runs of the algorithm', but it does not specify explicit train/validation/test dataset splits for model training or evaluation. |
| Hardware Specification | No | The paper does not provide specific hardware details (such as CPU or GPU models, processor types, or memory) used for running its experiments, only mentioning 'a Matlab implementation'. |
| Software Dependencies | No | The paper mentions 'a Matlab implementation of Perseus', but does not specify a version number for Matlab or any other software dependencies. |
| Experiment Setup | Yes | Our experiments have been performed with a Matlab implementation of Perseus (Spaan and Vlassis 2005) using basis functions to approximate value functions. All tests used a belief set of 500 belief points, and results are averaged over 10 times the number of start positions runs of the algorithm (we test all possible states as start positions). The benchmark problem domains for our tests are a variation of the fire fighting problem (Oliehoek et al. 2008) and the network management problem (Poupart 2005). |