Fast Model Identification via Physics Engines for Data-Efficient Policy Search

Authors: Shaojun Zhu, Andrew Kimmel, Kostas E. Bekris, Abdeslam Boularias

IJCAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Evaluations, performed both in simulation and on a real robotic manipulation task, indicate that the proposed strategy results in an overall timeefficient, integrated model identification and learning solution, which significantly improves the dataefficiency of existing policy search algorithms.
Researcher Affiliation Academia Shaojun Zhu, Andrew Kimmel, Kostas E. Bekris, Abdeslam Boularias Department of Computer Science, Rutgers University, New Jersey, USA
Pseudocode Yes Algorithm 1: Main Loop
Open Source Code No The paper mentions the use of third-party open-source tools (MuJoCo, OpenAI Gym, rllab) but does not state that the code for the specific methodology presented in this paper is open-source or provide a link to it.
Open Datasets Yes The simulation experiments are performed in Open AI Gym [Brockman et al., 2016] with the Mu Jo Co simulator1.
Dataset Splits No The paper does not provide specific percentages, sample counts, or explicit descriptions of how the data was split into training, validation, and test sets.
Hardware Specification No The paper mentions using specific robots (Motoman, Baxter) and a simulator (MuJoCo) but does not provide any details about the computational hardware (e.g., CPU, GPU models, memory) used for running the simulations or training.
Software Dependencies No The paper names software components like Open AI Gym, MuJoCo, and rllab, but it does not provide specific version numbers for any of them.
Experiment Setup Yes The policy network has 2 hidden layers with 32 neurons each.