Learning Safe Policies with Expert Guidance

Authors: Jessie Huang, Fa Wu, Doina Precup, Yang Cai

NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our experiments demonstrate the behavior of our algorithm in both discrete and continuous problems.
Researcher Affiliation Collaboration Jessie Huang1 Fa Wu12 Doina Precup1 Yang Cai1 1School of Computer Science, Mc Gill University 2Zhejiang Demetics Medical Technology
Pseudocode Yes Algorithm 1 Separation Oracle SOR for the reward polytope PR; Algorithm 2 Separation Oracle for the feasible (µ, z) in LP 1; Algorithm 3 FPL Maxmin Learning
Open Source Code No The paper does not provide any explicit statements about open-source code availability or links to code repositories for the described methodology.
Open Datasets Yes Our next experiments are based on the classic control task of cartpole and the environment provided by Open AI Gym [6].
Dataset Splits No The paper does not specify explicit training, validation, or test dataset splits in terms of percentages or sample counts. It describes using a 'small (10x10) demonstration gridworld' for expert policy generation and then testing in a 'much larger size (50x50)', but no specific splits are given.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments, such as GPU/CPU models, memory, or specific computing environments.
Software Dependencies No The paper does not provide specific version numbers for any software dependencies or libraries used in the experiments.
Experiment Setup Yes In the following experiment we set = 0.5 which defines PR and captures how close to optimal the expert is.