OFFER: Off-Environment Reinforcement Learning

Authors: Kamil Ciosek, Shimon Whiteson

AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we empirically compare OFFER to a policy gradient baseline (primary optimisation only) on variants of the mountain car task, as well as a simulated robot arm.
Researcher Affiliation Academia Kamil Ciosek, Shimon Whiteson Department of Computer Science, University of Oxford, United Kingdom {kamil.ciosek,shimon.whiteson}@cs.ox.ac.uk
Pseudocode Yes Algorithm 1 CRITIC-REINFORCE(τ) ... Algorithm 5 SECONDARY-OPTIMISATION(τ, ψ)
Open Source Code No The paper does not provide any statement or link indicating the release of open-source code for the described methodology.
Open Datasets No The paper mentions the 'mountain car benchmark task' and a 'simulated robotic-arm control task' proposed by Paul et al. (2016) but does not provide concrete access information (link, DOI, repository, or clear citation to a publicly available dataset) for the specific datasets or environments used in their modified experiments.
Dataset Splits No The paper does not explicitly state specific training, validation, or test dataset splits or percentages.
Hardware Specification No The paper does not provide any specific hardware details such as GPU or CPU models, memory, or cluster specifications used for running its experiments.
Software Dependencies No The paper mentions algorithms like ADAM but does not provide specific software names with version numbers for reproducibility (e.g., 'Python 3.x', 'PyTorch 1.x').
Experiment Setup No The paper states 'complete details are in the supplementary material' for the experimental setup, and the main text only provides general descriptions like 'θ consists of the parameters of a standard tile-coding function approximator' and 'the agent moves each joint by at most 30% of the allowed movement range', without specific hyperparameter values or comprehensive system-level training settings.