State Sequences Prediction via Fourier Transform for Representation Learning

Authors: Mingxuan Ye, Yufei Kuang, Jie Wang, Yang Rui, Wengang Zhou, Houqiang Li, Feng Wu

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments demonstrate that the proposed method outperforms several state-of-the-art algorithms in terms of both sample efficiency and performance.2
Researcher Affiliation Academia 1CAS Key Laboratory of Technology in GIPAS, University of Science and Technology of China 2Institute of Artificial Intelligence, Hefei Comprehensive National Science Center {mingxuanye, yfkuang, yr0013}@mail.ustc.edu.cn {jiewangx, zhwg, lihq, fengwu}@ustc.edu.cn
Pseudocode Yes The algorithm pseudo-code is shown in Appendix B. Algorithm 1 State Sequences Prediction via Fourier Transform (SPF)
Open Source Code Yes 2The code of SPF is available on Git Hub at https://github.com/MIRALab-USTC/RL-SPF/. and Codes for the proposed method are available at https://github.com/MIRALab-USTC/RL-SPF/.
Open Datasets Yes We quantitatively evaluate our method on a standard continuous control benchmark the set of Mu Jo Co [27] environments implemented in Open AI Gym. and citation [27] refers to Emanuel Todorov, Tom Erez, and Yuval Tassa. Mujoco: A physics engine for model-based control. In International Conference on Intelligent Robots and Systems, IROS, pages 5026 5033. IEEE, 2012.
Dataset Splits No The paper evaluates performance based on 'steps' (e.g., 500K step and 1M step) for 'sample efficiency' but does not specify explicit training, validation, and test dataset splits or a cross-validation setup for reproducibility of data partitioning.
Hardware Specification No The paper does not provide specific details about the hardware (e.g., GPU models, CPU types, or cloud computing specifications) used to run the experiments.
Software Dependencies No The paper mentions software like 'Open AI Gym', 'Mu Jo Co', 'SAC', and 'PPO', but does not provide specific version numbers for these or any other software dependencies, which are necessary for reproducible descriptions.
Experiment Setup Yes Table 3: Hyperparameters of auxiliary prediction tasks. Hyperparameters of SPF-SAC. Hyperparameters of SPF-PPO. This table specifies optimizer, learning rate, batch size, network architecture details, and various training parameters.