Context-Specific Representation Abstraction for Deep Option Learning

Authors: Marwa Abdulhai, Dong-Ki Kim, Matthew Riemer, Miao Liu, Gerald Tesauro, Jonathan P. How5959-5967

AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We test our method against hierarchical, nonhierarchical, and modular recurrent neural network baselines, demonstrating significant sample efficiency improvements in challenging partially observable environments. Our experiments demonstrate how CRADOL can improve performance over key baselines in partially observable settings.
Researcher Affiliation Collaboration 1 MIT LIDS 2 MIT-IBM Watson AI Lab 3 IBM Research abdulhai@mit.edu, dkkim93@mit.edu, mdriemer@us.ibm.com, miao.liu1@ibm.com, gtesauro@us.ibm.com, jhow@mit.edu
Pseudocode No The paper does not contain any sections or figures explicitly labeled "Pseudocode" or "Algorithm" with structured code-like steps.
Open Source Code Yes The code is available at https://git.io/JucVH
Open Datasets Yes Mini Grid (Chevalier-Boisvert, Willems, and Pal 2018): A library of open-source grid-world domains in sparse reward settings with image observations. Moving Bandit (Frans et al. 2017): This 2D sparse reward setting considers a number of marked positions in the environment... Reacher (Brockman et al. 2016): In this simulated Mu Jo Co task of Open AI Gym environment...
Dataset Splits No The paper does not explicitly provide details about training, validation, and test dataset splits with percentages or counts. It mentions
Hardware Specification No The paper does not specify any hardware details such as CPU, GPU models, or memory used for running the experiments.
Software Dependencies No The paper mentions software environments like "Mini Grid (Chevalier-Boisvert, Willems, and Pal 2018)", "Open AI Gym environment", and "Mu Jo Co task", but does not provide specific version numbers for these or other software dependencies.
Experiment Setup No The paper states: "We explain further details on experimental settings, including domains and hyperparameters, in the appendix." This indicates that specific experimental setup details like hyperparameters are not present in the main text.