Tool Use Baselines

Experiment 1

python -m tool_use.main --job-dir ~/jobs/tool_use/(date +"%s") --env-name 'TrapTube-v0'

Experiment 2

python -m tool_use.main --job-dir ~/jobs/tool_use/(date +"%s") --env-name 'PerceptualStructuralSymbolicTrapTube-v0'

Citation

If you use this code in your work, please cite the following:

@ARTICLE{2019arXiv190702050W,
       author = {{Wenke}, Sam and {Saunders}, Dan and {Qiu}, Mike and {Fleming}, Jim},
        title = "{Reasoning and Generalization in RL: A Tool Use Perspective}",
      journal = {arXiv e-prints},
     keywords = {Computer Science - Neural and Evolutionary Computing, Computer Science - Artificial Intelligence, Computer Science - Machine Learning},
         year = "2019",
        month = "Jul",
          eid = {arXiv:1907.02050},
        pages = {arXiv:1907.02050},
archivePrefix = {arXiv},
       eprint = {1907.02050},
 primaryClass = {cs.NE},
       adsurl = {https://ui.adsabs.harvard.edu/abs/2019arXiv190702050W},
      adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}