Multi-Modal Imitation Learning from Unstructured Demonstrations using Generative Adversarial Nets

Authors: Karol Hausman, Yevgen Chebotar, Stefan Schaal, Gaurav Sukhatme, Joseph J. Lim

NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The extensive simulation results indicate that our method can efficiently separate the demonstrations into individual skills and learn to imitate them using a single multi-modal policy. The video of our experiments is available at http://sites.google.com/view/nips17intentiongan.
Researcher Affiliation Academia Karol Hausman , Yevgen Chebotar , Stefan Schaal , Gaurav Sukhatme , Joseph J. Lim University of Southern California, Los Angeles, CA, USA Max-Planck-Institute for Intelligent Systems, Tübingen, Germany {hausman, ychebota, sschaal, gaurav, limjj}@usc.edu
Pseudocode No The paper describes the mathematical formulation of the proposed method but does not include any pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any link to open-source code for the described methodology. It only links to a video of experiments.
Open Datasets No The paper describes that demonstrations are used to train the multi-modal policy and that expert policies were used to create the combined dataset, but it does not provide concrete access information (link, DOI, citation) to a publicly available or open dataset.
Dataset Splits No The paper does not provide specific details on training, validation, or test dataset splits (e.g., percentages, sample counts, or citations to predefined splits).
Hardware Specification No The paper mentions 'simulation results' and 'simulated robotics tasks' but does not specify any hardware details (e.g., CPU, GPU models) used for running the simulations or training the models.
Software Dependencies No The paper mentions various frameworks and techniques like 'neural networks', 'TRPO', 'GANs', but it does not provide specific software names with version numbers (e.g., Python 3.x, TensorFlow x.y).
Experiment Setup Yes During our experiments, we anneal the noise similar to [32], as the generator policy improves towards the end of the training. For the continuous latent variable, we show a span of different intentions between -1 and 1 in the 0.2 intervals.