Provable Representation Learning for Imitation with Contrastive Fourier Features

Authors: Ofir Nachum, Mengjiao Yang

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on both tabular environments and high-dimensional Atari games provide quantitative evidence for the practical benefits of our proposed objective.
Researcher Affiliation Industry Ofir Nachum Google Brain ofirnachum@google.com Mengjiao Yang Google Brain sherryy@google.com
Pseudocode Yes see Appendix A for pseudocode.
Open Source Code Yes Find experimental code at https://github.com/google-research/google-research/tree/ master/rl_repr.
Open Datasets Yes We take for the offline dataset the DQN Replay Dataset [7], which for each game provides 50M steps collected during DQN training.
Dataset Splits No The paper describes the datasets used for training and evaluation but does not specify explicit train/validation/test splits for reproducibility.
Hardware Specification No The paper does not explicitly describe the specific hardware (e.g., GPU models, CPU types, or memory) used for running experiments.
Software Dependencies No The paper mentions extending implementations from 'Dopamine [14]' but does not provide specific version numbers for any software dependencies.
Experiment Setup Yes In our implementation we choose αR = 1, αT = (1 γ) 1 to roughly match the coefficients of the bounds in Theorems 2 and 3.