Measuring Structural Similarities in Finite MDPs

Authors: Hao Wang, Shaokang Dong, Ling Shao

IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We experimentally evaluate our proposed similarity measures using a series of designed MDPs.
Researcher Affiliation Collaboration 1Inception Institute of Artificial Intelligence, UAE 2State Key Laboratory for Novel Software Technology, Nanjing University, China
Pseudocode Yes Algorithm 1 shows the iterative algorithm for computing σ S and σ A by simulating the recursion of Sec. 4.1.
Open Source Code No The paper does not mention or provide any links to open-source code for the described methodology.
Open Datasets No The paper states, 'We generate a series of n n grid MDPs {Mn}...'. This indicates a self-generated dataset for which no public access information (link, DOI, citation) is provided.
Dataset Splits No The paper describes experiments on generated MDPs but does not specify any training, validation, or test dataset splits.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used to run the experiments.
Software Dependencies No The paper mentions algorithms like 'Dijkstra’s algorithm' and 'successive shortest path (SSP) algorithm' but does not specify any software dependencies with version numbers (e.g., Python 3.x, PyTorch 1.x).
Experiment Setup Yes In all the experiments, we set CS = CA = 0.95 for our solution and c R = 0.05 and c T = 0.95 for dbis. ... Algorithm 1 calculates two similarity matrices, Sn and An, with parameters CS = CA = 0.95... We first fix CS = 0.95 and vary CA from 0.80 to 0.99 with a step-size of 0.01. Then, we swap the roles of CS and CA.