Adaptive Neural Network Architecture.

reportActive / Technical Report | Accession Number: ADA190114 | Open PDF

Abstract:

Sun microsystems computer equipment acquired through Grant AFOSR-86-0260 is being used for research directed toward developing learning methods and architectures for artificial neural networks, or connectionist networks. The equipment is being used to simulate artificial neural networks implementing a variety of learning methods, including the Associative Reward-Penalty method and the Adaptive Critic Algorithm, as well as the error backpropagation method, and various combinations of these learning methods.

Author(s):
Barto, Andrew
Author Organization(s):
MASSACHUSETTS UNIV AMHERST
Descriptive Note:
Final rept. 1 Aug 86-31 Jul 87,
Supplementary Note:
DOI: 10.21236/ADA190114
Pagination:
0004

Security Markings

DOCUMENT & CONTEXTUAL SUMMARY

Distribution:
Approved For Public Release

RECORD

Collection: TR
Identifying Numbers
Report Number(s):
AFOSR-TR-87-1789
Contract/Grant Number(s):
AFOSR-86-0260
Task Number(s):
A4
Project Number(s):
2917
Monitor Series:
TR-87-1789
 Subject Terms
Joint Capability Areas:
JCA_5.5.2_Task; JCA_5.5_Direct; JCA_5_Command and Control
Modernization Areas:
AI and Machine Learning
Communities of Interest:
Autonomy
Descriptor(s):
*NEURAL NETS, ADAPTIVE SYSTEMS, ALGORITHMS, LEARNING, COMPUTER ARCHITECTURE, COMMUNICATIONS NETWORKS, COMPUTER COMMUNICATIONS, RESEARCH MANAGEMENT
Field(s)/Group(s):
Non-Radio Communications, Computer Systems
Keyword(s):
PE61102F, WUAFOSR2917A4
Report Date:
1987 Oct 28