UMBC CS Colloq 2/19: Learning Probabilities for Causal Networks

Article View: dc.general

Article #55371

#55371

From: Timothy Finin
Date: Mon, 17 Feb 1992 22:49

19 lines
1953 bytes

To represent the causal associations more accurately, an innovative
new concept, called causation event was proposed in a previous AI
causal model.  As hidden events, probabilities of causations cannot be
obtained using ordinary statistical methods. Based on the AI causal
model, the basic learning rule is extended by including a stochastic
factor.  The resulting stochastic learning rule was shown to be
capable of learning probabilities of causations.

Computer simulations were performed for both the basic and the
stochastic learning rules.  The results will also be presented.
				   

		   Wednesday, February 19, 1:00 PM
		     Computer Science Department
		    Room 102 Temporary Facility 1
			  UMBC, Baltimore MD
				   
  For additional information, contact Angie Silanskis, 410-455-3000.

Technical Details

Message-ID: <1992Feb17.224930.6129@umbc3.umbc.edu>

Path: arc1-ca.pugleaf.net!sparky!uunet!haven.umd.edu!news.umbc.edu!algol.cs.umbc.edu!finin

Show Full Headers

Complete Article Headers

X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 104fe2,19c88290f918edb,start
X-Google-Attributes: gid104fe2,public
X-Google-Thread: 101049,19c88290f918edb,start
X-Google-Attributes: gid101049,public
X-Google-Thread: 120a9d,19c88290f918edb,start
X-Google-Attributes: gid120a9d,public
X-Google-ArrivalTime: 1992-02-17 15:02:10 PST
Newsgroups: dc.general,umbc.general,umbc.cs
Path: arc1-ca.pugleaf.net!sparky!uunet!haven.umd.edu!news.umbc.edu!algol.cs.umbc.edu!finin
From: finin@algol.cs.umbc.edu (Timothy Finin)
Subject: UMBC CS Colloq 2/19: Learning Probabilities for Causal Networks
Message-ID: <1992Feb17.224930.6129@umbc3.umbc.edu>
Sender: newspost@umbc3.umbc.edu (News posting account)
Organization: Computer Science, University of Maryland, Baltimore County
         COLLOQUIUM -- COLLOQUIUM
       Computer Science Department
        University of Maryland, Baltimore County
       LEARNING PROBABILITIES FOR CAUSAL NETWORKS
          Yun Peng
      Department of Computer Science
        University of Maryland Baltimore County
     ypeng@cs.umbc.edu
Distribution: dc
Date: Mon, 17 Feb 1992 22:49:30 GMT