Peter Stone's Selected Publications

Classified by TopicClassified by Publication TypeSorted by DateSorted by First Author Last NameClassified by Funding Source


Overlapping Layered Learning

Patrick MacAlpine and Peter Stone. Overlapping Layered Learning. Artificial Intelligence, 254:21–43, Elsevier, January 2018.
Official version from Publisher's Webpage
Accompanying videos at http://www.cs.utexas.edu/~AustinVilla/sim/3dsimulation/overlappingLayeredLearning.html

Download

[PDF]1.2MB  [postscript]3.8MB  

Abstract

Layered learning is a hierarchical machine learning paradigm thatenables learning of complex behaviors by incrementally learning a series of sub-behaviors. A key feature of layered learning is that higher layers directly depend on the learned lower layers. In its original formulation, lower layers were frozen prior to learning higher layers. This article considers a major extension to the paradigm that allows learning certain behaviors independently, and then later stitching them together by learning at the "seams" where their influences overlap. The UT Austin Villa 2014 RoboCup 3D simulation team, using such overlapping layered learning, learned a total of 19 layered behaviors for a simulated soccer-playing robot, organized both in series and in parallel. To the best of our knowledge this is more than three times the number of layered behaviors in any prior layered learning system. Furthermore, the complete learning process is repeated on four additional robot body types, showcasing its generality as a paradigm for efficient behavior learning. The resulting team won the RoboCup 2014 championship with an undefeated record, scoring 52 goals and conceding none. This article includes a detailed experimental analysis of the team's performance and the overlapping layered learning approach that led to its success.

BibTeX Entry

@article{AIJ18-MacAlpine,
  title = {Overlapping Layered Learning},
  journal = {Artificial Intelligence},
  volume = {254},
  pages = {21--43},
  month = {January},
  year = {2018},
  issn = {0004-3702},
  doi = {https://doi.org/10.1016/j.artint.2017.09.001},
  url = {https://www.sciencedirect.com/science/article/pii/S0004370217301066},
  author = {Patrick MacAlpine and Peter Stone},
  publisher = {Elsevier},
  abstract = {Layered learning is a hierarchical machine learning paradigm that
enables learning of complex behaviors by incrementally learning a series of 
sub-behaviors. A key feature of layered learning is that higher layers directly 
depend on the learned lower layers. In its original formulation, lower layers 
were frozen prior to learning higher layers. This article considers a major 
extension to the paradigm that allows learning certain behaviors independently, 
and then later stitching them together by learning at the "seams" where their 
influences overlap. The UT Austin Villa 2014 RoboCup 3D simulation team, using 
such overlapping layered learning, learned a total of 19 layered behaviors for 
a simulated soccer-playing robot, organized both in series and in parallel. To 
the best of our knowledge this is more than three times the number of layered 
behaviors in any prior layered learning system. Furthermore, the complete 
learning process is repeated on four additional robot body types, showcasing 
its generality as a paradigm for efficient behavior learning. The resulting 
team won the RoboCup 2014 championship with an undefeated record, scoring 52 
goals and conceding none. This article includes a detailed experimental 
analysis of the team's performance and the overlapping layered learning 
approach that led to its success.},
  wwwnote = {Official version from <a href="https://authors.elsevier.com/a/1Vtn--c5F-HA">Publisher's Webpage</a><br>
Accompanying videos at <a href="http://www.cs.utexas.edu/~AustinVilla/sim/3dsimulation/overlappingLayeredLearning.html">http://www.cs.utexas.edu/~AustinVilla/sim/3dsimulation/overlappingLayeredLearning.html</a>}
}

Generated by bib2html.pl (written by Patrick Riley ) on Fri Jun 14, 2019 11:36:24