Ibc

Implicit Behavioral Cloning - PyTorch

Pytorch implementation of <a href="https://arxiv.org/abs/2109.00137">Implicit Behavioral Cloning</a>.

Install

conda create -n ibc python=3.8
pip install -r requirements.txt

Results

To reproduce results from the Coordinate Regression Task (Section 3), execute the run_explicit.sh and run_implicit.sh scripts. Note that the implicit policy does a tad bit worst with 30 examples than 10. Not entirely sure why that is the case and need to investigate more.

| | Explicit Policy | Implicit Policy | |-------------|-----------------|-----------------| | 10 examples |<img src="assets/explicit_mse_10.png" width="300" height="200"/>|<img src="assets/implicit_ebm_10.png" width="300" height="200"/>| | 30 examples |<img src="assets/explicit_mse_30.png" width="300" height="200"/>|<img src="assets/implicit_ebm_30.png" width="300" height="200"/>|

Citation

If you find this code useful, consider citing it along with the paper:

@software{zakka2021ibc,
    author = {Zakka, Kevin},
    month = {10},
    title = {{A PyTorch Implementation of Implicit Behavioral Cloning}},
    url = {https://github.com/kevinzakka/ibc},
    version = {0.0.1},
    year = {2021}
}

@misc{florence2021implicit,
    title = {Implicit Behavioral Cloning},
    author = {Pete Florence and Corey Lynch and Andy Zeng and Oscar Ramirez and Ayzaan Wahid and Laura Downs and Adrian Wong and Johnny Lee and Igor Mordatch and Jonathan Tompson},
    year = {2021},
    eprint = {2109.00137},
    archivePrefix = {arXiv},
    primaryClass = {cs.RO}
}