http:^^envy.cs.umass.edu^people^barto^barto.html

来自「This data set contains WWW-pages collect」· HTML 代码 · 共 484 行 · 第 1/2 页

Date: Thu, 21 Nov 1996 20:49:21 GMT
Server: NCSA/1.5.2
Last-modified: Tue, 03 Sep 1996 18:37:53 GMT
Content-type: text/html
Content-length: 22052

<title>
Andrew G. Barto
</title>

<h1>
Andrew G. Barto
</h1>

<!WA0><img src="http://envy.cs.umass.edu/People/barto/barto.gif">

<P>
<STRONG>
Professor<BR>
<!WA1><A HREF="http://www.cs.umass.edu/">Department of Computer Science</A><BR>
<!WA2><A HREF="http://www.cs.umass.edu/rcfdocs/newhome/">University of Massachusetts</a>
<BR>
<!WA3><A HREF="http://www-astro.phast.umass.edu/guest/amherst.html">Amherst</A>, MA
01003  USA<P>

Co-director, <!WA4><A HREF="http://envy.cs.umass.edu/">Adaptive NetWork Laboratory</A
><BR>
</STRONG>
<BR>

<!WA5><A HREF="mailto:[e-mail]">[e-mail]</A><BR>
Office: Lederle A265<BR>
413-545-2109, fax:[phone]

<UL>
<LI><!WA6><A HREF="#Bio">Brief Biography</A>
<LI><!WA7><A HREF="#Ph.D.">Ph.D. Students Advised</A>
<LI> Publications
	<UL>
		<LI> <!WA8><A HREF="#Refereed">Refereed Publications</A>
		<LI> <!WA9><A HREF="#Chapters">Book Chapters</A>
		<LI> <!WA10><A HREF="#Other">Other Publications</A>
	</UL>
</UL> <BR>

My research interests center on machine and biological learning. I have been
trying to develop learning algorithms that are useful for engineering
applications while also making contact with learning as studied by experimental
psychologists and neuroscientists. I am interested in artificial and real
neural networks, and over the last several years I have focused on connections
between reinforcement learning algorithms and dynamic programming solutions to
Markov decision problems. Related research is being conducted in collaboration
with colleagues specializing in animal motor control. We are working on a model
of the cerebellum and other brain regions involved in motor control.
<BR>  <HR>

<H2><A Name = "Bio">Brief Biography</A></H2>

B.S. with distinction in mathematics, 1970, University of Michigan; Ph.D. in
Computer Science, 1975, University of Michigan. From 1975 to 1979 he was an
Assistant Professor at the School of Advanced Technology, SUNY, Binghamton, NY.
Taking a leave of absence in 1977, he became a Postdoctoral Research Associate
in the Computer and Information Science Department, 
University of Massachusetts, Amherst, where he was appointed Associate
Professor in 1982. In 1991 he was promoted to his current position,
Professor of Computer Science, University of Massachusetts, Amherst.
In addition to co-directing the Adaptive NetWork Laboratory, he is a core
faculty of the Neuroscience and Behavior Program,  University of Massachusetts.
Member: Society for Neuroscience and INNS; senior member of the IEEE;
member and fellow of the American Association for the Advancement of
Science. INNS board of governors from 1991-95. Member 
editorial board and action editor, <cite>Neural Networks</cite>, 1987-95. Action
editor, <cite>Machine Learning</cite>; associate editor, <cite>Neural
Computation</cite>; member editorial board, <cite>Journal of Artificial
Intelligence Research</cite>; associate editor, MIT Press book series on Neural
Network Modeling and Connectionism.   

<hr>

<H2><A Name = "Ph.D.">Ph.D. Students Advised</A></H2>
<P>
1984: <!WA11><A HREF="http://envy.cs.umass.edu/People/sutton/sutton.html">
R. S. Sutton</A>, ``Temporal Credit Assignment in Reinforcement
		Learning.&quot;
<P>
1986: C. W. Anderson, ``Learning and Problem Solving with Multilayer
		Connectionist Systems.&quot;
<P>
1988: <!WA12><A HREF="http://envy.cs.umass.edu/cgi-bin/finger?[e-mail]">J. S. Judd</A>,
``Neural Network Design and the Complexity of Learning.&quot; 
<P> 
1990: <!WA13><A HREF="http://envy.cs.umass.edu/cgi-bin/finger?[e-mail]">R. A. Jacobs</A>,
``Task Decomposition through Competition in a Modular Connectionist
Architecture.&quot;
<P> 
1992: J. R. Bachrach, ``Connectionist Modeling and
Control of Finite State Environments.&quot;
<P>
1992: <!WA14><A HREF="http://envy.cs.umass.edu/People/vijay/vijay.html">V. Gullapalli</A>,
``Reinforcement Learning and Its Application to Control.&quot;
<P>
1993: <!WA15><A HREF="http://envy.cs.umass.edu/People/singh/singh.html">S. P. Singh</A>,  ``Learning
to Solve Markovian Decision Processes.&quot;
<P>
1994: <!WA16><A HREF="http://envy.cs.umass.edu/People/bradtke/bradtke.html">S. J. Bradtke</A>,  ``Incremental
Dynamic Programming for On-Line Adaptive Optimal Control.&quot;
<hr>

<H2>PUBLICATIONS</H2>

<H3><A NAME = "Refereed">Refereed Publications</A></H3>
<P>
R. Crites and A. Barto.
<!WA17><a href="http://envy.cs.umass.edu/cgi-bin/getfile/pub/anw/pub/crites/nips8.ps.Z">
<i>Improving Elevator Performance Using Reinforcement Learning</i>
</a>. To appear in <i>Advances in Neural Information Processing Systems 8</i>
(NIPS8). (nips8.ps.Z: 58525 bytes)
<p>
R. Crites and A. Barto. 
<!WA18><a href="http://envy.cs.umass.edu/cgi-bin/getfile/pub/anw/pub/crites/nips7.ps.Z">
<i>An Actor/Critic Algorithm that is Equivalent to Q-Learning</i>
</a>. <i>Advances in Neural Information Processing Systems 7</i> (NIPS7),
G. Tesauro, D. S. Touretzky, and T. K. Leen (Eds.), Cambridge, MA: MIT Press,
1995, pp. 401-408. (nips7.ps.Z: 64695 bytes)
<P>
 A. G. Barto, S. J. Bradtke and S. P. Singh.  ``Learning to act
using real-time dynamic programming.&quot  <i>Artificial Intelligence</i>,
Special Volume: Computational Research on Interaction and Agency, <b>72</b>(1):
81-138, 1995.
<P>
S. P. Singh, A. G. Barto, R. Grupen and C. Connolly. ``Robust Reinforcement Learning in Motion
Planning.&quot <i>Neural Information Processing
Systems 6</i> (NIPS6),  J. D. Cowan, G. Tesauro, and J. Alspector (Eds.),
San Mateo: Morgan Kaufmann, 1994, pp. 655-662.
<P>
V. Gullapalli, A. G.  Barto and R. A. Grupen.  ``Learning admittance
mappings for force-guided assembly.&quot  <i>Proceedings of the 1994 
International Conference on Robotics and Automation,</i>1994, pp. 2633-2638.
<P>
V. Gullapalli and A. G. Barto. ``Convergence of Indirect Adaptive
Value Iteration.&quot <i>Neural Information Processing
Systems 6</i> (NIPS6), J. D. Cowan, G. Tesauro, and J. Alspector (Eds.), San
Mateo: Morgan Kaufmann, 1994, pp. 695-662.
<P>
J. T. Buckingham, J. C. Houk, and A. G. Barto. ``Controlling a Nonlinear Spring-Mass System with a
Cerebellar Model.&quot <i>Proceedings of the 8th Yale Workshop on Adaptive and
Learning Systems</i>. Yale University, 1994, pp. 1-6.
<P>
S. J. Bradtke, A. G. Barto and B. E. Ydstie. ``A Reinforcement Learning Method for Direct 
Adaptive Linear Quadratic Control.&quot <i>Proceedings of the 8th Yale Workshop
on Adaptive and Learning Systems</i>. Yale University, 1994, pp. 85-96.
<P>
A. G. Barto and M. Duff. ``Monte-Carlo Matrix Inversion and
Reinforcement Learning.&quot
<i>Neural Information Processing
Systems 6</i> (NIPS6),  J. D. Cowan, G. Tesauro, and J. Alspector (Eds.),
San Mateo: Morgan Kaufmann, 1994, pp. 687-662.
<P>
J. C. Houk, J. Kiefer, and A. G. Barto.  ``Distributed motor
commands in the limb premotor network.&quot  <i>Trends in Neuroscience,</i>
<b>16</b> (1): 27-33, 1993.
<P>
N.E. Berthier, S.P. Singh, A.G. Barto, and J.C. Houk. ``Distributed Representations of Limb Motor
Programs in Arrays of Adjustable Pattern Generators&quot <i>Journal of
Cognitive Neuroscience</i>, <b>5</b> (1): 56-78, 1993.
<P>
V. Gullapalli, R. A. Grupen, and A. G. Barto. 
``Learning Reactive Admittance Control.&quot
<i>Proceedings of the 1992 IEEE International Conference on Robotics and
 Automation.</i> Nice, France, May 1992,
pp. 1475-1480.
<P>
V. Gullapalli and A. G. Barto. ``Shaping as a Method for
 Accelerating Reinforcement Learning.&quot
<i>Proceedings of the 1992 IEEE International Symposium on Intelligent
Control</i>. Glasgow, Scotland, August 1992, pp. 554-559.
<P>
N.E. Berthier, S.P. Singh, A.G. Barto, and J.C. Houk. ``A Cortico-Cerebellar Model that Learns to
Generate Distributed Motor Commands to Control a Kinematic Arm.&quot <i>Neural
Information Processing Systems 4</i> (NIPS4), J. E. Moody, S. J. Hanson, and R.
P. Lippmann (Eds.). San Mateo: Morgan Kaufmann, 1992, pp. 611-618.
<P>
A. G. Barto and S. J. Bradtke. ``Learning to Solve Stochastic Shortest Path Problems using
Real-Time Dynamic Programming.&quot <i>Proceedings of the Seventh Yale Workshop
on Adaptive and Learning Systems</i>. New Haven CT, 1992, pp. 143-148.
<P>
N. Berthier, A. Barto, and J. Moore. ``Linear systems analysis of the
relationship between firing of deep cerebellar neurons and the classically
conditioned nictitating membrane response in rabbits.&quot <i>Biological
Cybernetics</i>, <b>65</b>: 99-105, 1991.
<P>
A.G. Barto and S.P. Singh, 
 ``Reinforcement learning and dynamic programming,&quot 
<i>Proceedings of the Sixth Yale Workshop on Adaptive and Learning Systems</i>.
New Haven, CT, 1990, pp. 83-88.
<P>
R.A. Jacobs, M.I. Jordan and A.G. Barto, 
``Task Decomposition Through Competition in a Modular Connectionist 
Architecture: The What and Where Vision Task,&quot   <i>Cognitive Science</i>,
<b>15</b>:  219-250, 1991.
<P>
R.C. Yee, S. Saxena, P.E. Utgoff and A.G. Barto, 
``Explaining temporal differences to create useful concepts for evaluating 
states,&quot  <i>Proceedings of the 
Eighth National Conference on Artificial Intelligence</i>.
Cambridge, MA, August 1990, pp. 882-888.
<P>
T. Sinkj&#230;r, C.H. Wu, A.G. Barto,  and J.C. Houk,
``Cerebellar control of endpoint position - A simulation model,&quot 
 <i>Proceedings of the 1990 International
Joint Conference on Neural Networks</i>. San Diego, CA, June 1990, pp.
II-705-II-710. 
<P>
A.G. Barto,  R.S. Sutton and C. Watkins, 
 ``Sequential decision problems and neural networks,&quot  <i>Advances in Neural
Information Processing 2</i> (NIPS2),  D. Touretzky (Ed.). San Mateo, CA:	Morgan
Kaufmann, 1990, pp. 686-693.
<P>
R.S. Sutton and A.G. Barto, ``A temporal-difference model of classical
conditioning,&quot  <i>Proceedings of the Ninth Annual Conference
of the Cognitive Science Society</i>. Hillsdale, NJ: Erlbaum, 1987. 
<P>
A.G. Barto and M.I. Jordan, ``Gradient following without back-propagation in
layered networks,&quot  <i>Proceedings of the IEEE First Annual
Conference on Neural Networks</i>. San Diego, CA, June 1987, pp. II-629-II-636.
<P>
A.G. Barto, ``Game-theoretic cooperativity in networks of self-interested units,&quot;
in <i>Neural Networks for Computing</i>. J.S. Denker (Ed.). New
York: American Institute of Physics, 1986, pp. 41-46.
<P>
A.G. Barto, ``Learning by statistical cooperation of self-interested
neuron-like computing elements,&quot;
<i>Human Neurobiology</i>, <b>4</b>:  219-250, 1985.
<P>
J.W. Moore, J.E. Desmond, N.E. Berthier, D.E.J. Blazis, R.S. Sutton
and A.G. Barto,  
``Connectionistic learning in real time: Sutton-Barto adaptive
element and classical conditioning of the nictitating membrane response,&quot
<i>Proceedings of the Seventh Annual Conference of the Cognitive Science
Society</i>. Irvine, CA, August 1985.
<P>
O. Selfridge, R.S. Sutton and A.G. Barto, ``Training and tracking in