Identifiability in inverse reinforcement learning

Authors: Haoyang Cao, Samuel Cohen, Lukasz Szpruch

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The code and instructions was included in the supplemental material. We did not work with any particular dataset. ... For the purpose of demonstration, instead of reporting error bars, we included training and test errors with respect to each random seed. ... All the experiments have been conducted on Google Research Colaboratory with GPU as hardware accelerator. The experiment in Section 4 took 2 minutes and 14 seconds in total for all 6 random initializations.
Researcher Affiliation Academia Haoyang Cao Alan Turing Institute hcao@turing.ac.uk Samuel N. Cohen Mathematical Institute, University of Oxford and Alan Turing Institute samuel.cohen@maths.ox.ac.uk Łukasz Szpruch School of Mathematics, University of Edinburgh and Alan Turing Institute L.Szpruch@ed.ac.uk
Pseudocode No The paper does not contain any clearly labeled pseudocode or algorithm blocks.
Open Source Code Yes The code and instructions was included in the supplemental material.
Open Datasets No We did not work with any particular dataset.
Dataset Splits No The paper states, "We did not work with any particular dataset," which implies no dataset splits, including validation splits, were used for the experiments described.
Hardware Specification No All the experiments have been conducted on Google Research Colaboratory with GPU as hardware accelerator.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup No The ethics statement claims "We specified training details in the main text" and "The experiment in Section 4 took 2 minutes and 14 seconds in total for all 6 random initializations." However, the provided main text of the paper does not contain specific experimental setup details, hyperparameters, or training configurations for these numerical demonstrations.