Variational Adversarial Kernel Learned Imitation Learning
Authors: Fan Yang, Alina Vereshchaka, Yufan Zhou, Changyou Chen, Wen Dong6599-6606
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the performance of our algorithm through benchmarking with four state-of-the-art imitation learning algorithms over five high-dimensional control tasks, and a complex transportation control task. Experimental results indicate that our algorithm significantly outperforms related algorithms in all scenarios. |
| Researcher Affiliation | Academia | Fan Yang, Alina Vereshchaka, Yufan Zhou, Changyou Chen, Wen Dong State University of New York at Buffalo {fyang24, avereshc, yufanzho, changyou, wendong}@buffalo.edu |
| Pseudocode | Yes | Algorithm 1: Variational adversarial kernel learned imitation learning |
| Open Source Code | Yes | 1The appendix, codes and data used in this paper can be found in https://bit.ly/37rescf |
| Open Datasets | Yes | To demonstrate the performance of our algorithm, we first benchmark our algorithm against other state-of-the-art imitation learning algorithms over five high-dimensional control problems, integrated in Open AI Gym (Brockman et al. 2016): Ant, Half Cheetah, Humanoid, Humanoid Flagrun, and Walker2D (Schulman et al. 2017). We then test our algorithm on a more complicated traffic control problem in a transportation system (Yang, Liu, and Dong 2019). |
| Dataset Splits | No | The paper describes collecting expert trajectories and samples from the learned policy, and refers to 'benchmarking' and 'grid searches' for hyperparameter tuning. However, it does not explicitly provide specific training, validation, and test dataset splits with percentages or counts. |
| Hardware Specification | No | The paper does not provide any specific hardware details such as GPU/CPU models, processor types, or memory specifications used for running the experiments. |
| Software Dependencies | No | The paper mentions software like 'Open AI Gym' and 'TRPO' but does not provide specific version numbers for any software dependencies. |
| Experiment Setup | No | The paper states 'Detailed descriptions of the environments and the experimental setup, and additional experiments are given in the Appendix', implying that specific setup details are not present in the main text provided. |