Active Learning of Parameterized Skills

Authors: Bruno Da Silva, George Konidaris, Andrew Barto

ICML 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate it on a non-linear simulated catapult control problem over arbitrarily mountainous terrains.
Researcher Affiliation Academia Bruno Castro da Silva BSILVA@CS.UMASS.EDU School of Computer Science, University of Massachusetts Amherst, MA 01003 George Konidaris GDK@CSAIL.MIT.EDU Computer Science and Artificial Intelligence Lab, MIT, Cambridge, MA 02139 Andrew Barto BARTO@CS.UMASS.EDU School of Computer Science, University of Massachusetts Amherst, MA 01003
Pseudocode No The paper includes mathematical equations and derivations but no pseudocode or algorithm blocks.
Open Source Code Yes Code will be made available at http://bitbucket. org/bsilvapoa/active_paramskill.
Open Datasets No The paper uses a simulated environment with randomly generated terrains and does not refer to a publicly available dataset with concrete access information.
Dataset Splits No The paper describes evaluating skill performance on "a set of novel tasks" and averaging over "50 randomly generated terrains" but does not specify explicit training, validation, or test dataset splits.
Hardware Specification No The paper does not provide any specific details about the hardware used for running the experiments.
Software Dependencies No The paper does not specify any software names with version numbers.
Experiment Setup No The paper describes the experimental setup in terms of the catapult domain and evaluation metrics but does not provide specific hyperparameters, training configurations, or system-level settings.