06-ComputerScience-pt2 - Introduction to Cognitive Science...

Info iconThis preview shows pages 1–11. Sign up to view the full content.

View Full Document Right Arrow Icon
    Introduction to Cognitive Science PSYCH 1102/COGST 1101/LING 1170/ PHIL 1910/CS 1710
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Computer Science Connectionism Philosophy                               Psychology                Neuroscience Linguistics Science Mind?
Background image of page 2
    Outline Problems for Physical Symbol Systems Letting the air out of the PSS Connectionism A newish approach to AI Syntax & Semantics in networks Fuzzy weights & dynamic activation Learning in networks The good, the bad and not really any ugly PSS vs PDP Computers can learn English too Problems for connectionism “It’s only a model”
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Problems for Physical Symbol Systems Letting the air out of the PSS
Background image of page 4
    Problems for Physical Symbol Systems Empirical Issues No GOFAI is as good at doing these things  as humans, or at least it does them  differently Perceptual problems (object recognition,  reading handwriting, listening to language)  are notoriously difficult to solve with PSS PSS require a large amount of “built-in”  information It is unclear that people (and other animals)  come with that
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Problems for Physical Symbol Systems Philosophical Issues “The more than one way to skin a cat  conundrum” Functionalism: end result is what matters Thus AI is fine if it works But does GOFAI tell us about human cognition? Can we build an AI that tries to do cognition like  humans do? Pattern of learning Representation of information Time course of processing
Background image of page 6
    Connectionism A newish approach to AI
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Connectionism Instead of working at the level of  abstraction of the PSS… …which doesn’t end up acting all that human work at a different level of analysis Connectionist models Computational model of large networks  individually dumb units spreading patterns of  activation Also called Parallel Distributed Processing  (PDP), neural networks, fuzzy networks, etc.
Background image of page 8
    Connectionism Advantages over GOFAI Neural Plausibility Soft Constraints Graceful degradation Learning from experience
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
    Connectionism Neural Plausibility Large network of dumb units, interconnected &  processing information through time If cognition occurs in humans because of the  structure of our brains, connectionist system has  better chance than GOFAI of real cognition Doesn’t represent individual neurons Doesn’t represent neurotransmitters etc.
Background image of page 10
Image of page 11
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 62

06-ComputerScience-pt2 - Introduction to Cognitive Science...

This preview shows document pages 1 - 11. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online