Explicable Policy Search

Authors: Ze Gong, Yu ("Tony") Zhang

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

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate EPS in a set of navigation domains with synthetic human models and in an autonomous driving domain with a user study. The results suggest that our method can generate explicable behaviors that reconcile task performance with human expectations intelligently and has real-world relevance in human-agent teaming domains.
Researcher Affiliation Academia Ze Gong Arizona State University Tempe, AZ 85281 zgong11@asu.edu Yu Zhang Arizona State University Tempe, AZ 85281 yzhan442@asu.edu
Pseudocode No The algorithm for EPS is in the appendix.
Open Source Code Yes Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [Yes] See supplemental material.
Open Datasets No The paper mentions synthetic human models and a simulated autonomous driving domain but does not provide concrete access information like URLs, DOIs, or specific citations for publicly available datasets.
Dataset Splits No The paper does not explicitly state training/validation/test dataset splits with percentages or sample counts in the main text.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments.
Software Dependencies No The paper mentions using Soft Actor Critic (SAC) [21] but does not provide specific version numbers for any software dependencies.
Experiment Setup No The paper mentions tuning a reconciliation parameter λ for EPS but does not provide a comprehensive list of hyperparameters or detailed system-level training settings in the main text. It refers to the appendix for more details on λ tuning but not general setup.