🚀 go-pugleaf

          The University of Western Ontario
            Department of Computer Science

                    COLLOQUIUM


Speaker: Professor Charles Ling, Western Ontario
Title:   Answering Connectionist Challenge
         A symbolic model for learning Past Tenses of English Verbs
Time:    Wednesday, October 14, 1992, at 2:00pm
Place:   MC 320
Abstract:
Is artificial neural network (ANN) a better learning model for
``complicated'' tasks done in the human brain than traditional
symbolic methods? What kind of tasks can be learned and generalized
better/worse using ANN than symbolic methods?
We give some answers to these questions, and provide a
case study -- learning past tenses of English verbs --
that connectionist models have been claimed successful as a benchmark.
A challenge is posed by MacWhinney and Leinbach (1991)
suggesting that no symbolic methods can perform as well as
their connectionist model. We design a general purpose Symbolic
Pattern Associator (SPA) based on ID3 (see Quinlan (1986))
that can learn and generalize from any patterns.
In comparison to the connectionist models, the SPA is able to
generalize much better to the past tenses of unseen regular and
irregular verbs. The SPA's performances are also more psychologically
realistic in many aspects. In addition, the SPA is able to represent
the acquired knowledge in production rules which are meaningful to
human observers, and which allow further generalization and integration
into other learning modules; while the connectionist model leaves
the acquired knowledge in a mysterious black-box.

This represents collaborative work with Steve Cherwen and Marin Marinov.

MacWhinney and Leinbach (1991). Implementations are not conceptualizations:
Revising the verb model. In Cognition, volume 40. Pages 121 - 157. 1991

Quinlan (1986). Induction of decision trees. In Machine Learning,
Volume 1. Pages 81 - 106. 1986.