Main.ConnectionismAndLOT History
Hide minor edits - Show changes to output
Changed lines 43-44 from:
* Micro-feature representations
to:
* [[http://www-personal.umich.edu/~lormand/phil/cogsci/clot_4.gif|Micro-feature representations]]
Changed line 10 from:
* ConnectionismFodorPylyshyn
to:
* [[ConnectionismFodorPylyshyn]]
Changed lines 2-3 from:
* [Required] Garson, James. Connectionism. ''The Stanford Encyclopedia of Philosophy''. stanford:connectionism.
* [Recommended] Gary F. Marcus (2001). ''The Algebraic Mind : Integrating Connectionism and Cognitive Science'' MIT Press. isbn:0262133792
to:
* Garson, James. Connectionism. ''The Stanford Encyclopedia of Philosophy''. stanford:connectionism.
* Gary F. Marcus (2001). ''The Algebraic Mind : Integrating Connectionism and Cognitive Science'' MIT Press. isbn:0262133792
Changed line 67 from:
to:
[[GodelNumberingToyExample]]
Added lines 8-11:
!Fodor and Pylyshyn
* ConnectionismFodorPylyshyn
Changed lines 63-64 from:
to:
Changed lines 30-31 from:
* Localist representation - one node for one meaning. Cannot deal with productivity.
to:
* Localist representation - one node for one meaning. Can it deal with systematicity and productivity?
Added line 5:
* van Gelder, T. J. (1999). [[http://www.arts.unimelb.edu.au/~tgelder/papers/MITDist.pdf|Distributed versus local representation]]. In R. Wilson & F. Keil (Eds.) ''The MIT Encyclopedia of Cognitive Sciences''. Cambridge MA: MIT Press, 236-8.
Changed lines 60-62 from:
Compare: "Where do zipped files come from?", "How can you change one file in a zip archive without changing others?"
to:
Compare: "Where do zip files come from?", "How can you change one file in a zip archive without changing others?"
Changed line 32 from:
to:
Changed lines 35-37 from:
||border=0 valign=top
||media:theta-network.png||
* Structured, distributed representations can implement compositional representations.[[<<]][[<<]][- van der Veldea & de Kamp. Neural blackboard architectures of combinatorial structures in cognition. ''Behavioral and Brain Sciences''-]||
to:
* Structured, distributed representations can implement compositional representations.
media:theta-network.png
[- van der Veldea & de Kamp. Neural blackboard architectures of combinatorial structures in cognition. ''Behavioral and Brain Sciences''-]
Changed lines 35-36 from:
||border=0
to:
||border=0 valign=top
||media:theta-network.png||
* Structured, distributed representations can implement compositional representations.[[<<]][[<<]][- van der Veldea & de Kamp. Neural blackboard architectures of combinatorial structures in cognition. ''Behavioral and Brain Sciences''-]||
Changed lines 35-36 from:
* Structured, distributed representations can implement compositional representations.
media:theta-network.png [- van der Veldea & de Kamp. Neural blackboard architectures of combinatorial structures in cognition. ''Behavioral and Brain Sciences''-]
to:
||border=0
||media:theta-network.png||* Structured, distributed representations can implement compositional representations.[[<<]][[<<]][- van der Veldea & de Kamp. Neural blackboard architectures of combinatorial structures in cognition. ''Behavioral and Brain Sciences''-]||
Changed line 36 from:
to:
media:theta-network.png [- van der Veldea & de Kamp. Neural blackboard architectures of combinatorial structures in cognition. ''Behavioral and Brain Sciences''-]
Changed lines 47-48 from:
to:
Changed lines 35-36 from:
* Structured, distributed representations can implement compositional representations - example
to:
* Structured, distributed representations can implement compositional representations.
media:theta-network.png
Changed lines 40-41 from:
to:
Changed line 55 from:
* Objection #2: Where do distributed representations come from? (e.g. [[NetworkArchitecture|RAAM]])
to:
* Objection #2: Where do distributed representations come from? (e.g. RSG model, [[NetworkArchitecture|RAAM]])
Changed lines 38-42 from:
Distributed memory in Ramsey, W., Stich, S. P., & Garon, J. (1991).
* network architecture media:rsg1.jpg
* representations media:rsg2.jpg
to:
!!Distributed memory in Ramsey, W., Stich, S. P., & Garon, J. (1991).
network architecture
media:rsg1.jpg
Input representations
media:rsg2.jpg
Changed lines 47-55 from:
* Distributed representations are powerful and useful. But can they explain cognition without LOT?
** Objection #1: Unstructured distributed representations cannot explain systematicity.
*** See Chalmers (1990) for a reply.
** Objection #2: Where do distributed representations come from? (e.g. [[NetworkArchitecture|RAAM]])
*** "Where do zipped files come from?"
** Objection #3: LOT needed to explain free transformation in central cognitive processes (e.g. conscious thoughts).
*** "How can you change one file in a zip archive without changing others?"
to:
Distributed representations are powerful and useful. But can they explain cognition without LOT?
* Objection #1: Unstructured distributed representations cannot explain systematicity.
** See Chalmers (1990) for a reply.
* Objection #2: Where do distributed representations come from? (e.g. [[NetworkArchitecture|RAAM]])
* Objection #3: LOT needed to explain free transformation in central cognitive processes (e.g. conscious thoughts).
Compare: "Where do zipped files come from?", "How can you change one file in a zip archive without changing others?"
Added line 51:
*** "Where do zipped files come from?"
Changed lines 53-54 from:
*** An analogy with zip files.
to:
*** "How can you change one file in a zip archive without changing others?"
Deleted lines 37-41:
!!Re scenario #3
* Connectionists opt for networks that employ unstructured, distributed representations.
* These representations are not compositional, and hence not classical. The key issue is whether networks with such representations can explain systematicity and productivity.
* Some such networks are systematic and productive to a limited extent. But can they still perform satisfactorily when they are scaled up?
Added line 49:
*** See Chalmers (1990) for a reply.
Changed line 57 from:
to:
* Is this an example of LOT?
Changed lines 60-71 from:
*
[[Category.Mind]]
!An alternative to LOT?
Single-route vs dual-route theories. Past-tense network in Rumelhart and McClelland (1986).
Pinker : two mechanisms - one for regular verbs ("talk") and one for irregulars ("go").
Plunkett, Rumelhart, ... : just one single network for both verb classes.
* Recall the three levels : the task level which describes what the system can do, the computational level which describes the representations and computations used, the hardware level which is about how the computations are realized and constructed.
to:
Changed line 54 from:
** Objection #2: Where do distributed representations come from?
to:
** Objection #2: Where do distributed representations come from? (e.g. [[NetworkArchitecture|RAAM]])
Changed line 64 from:
to:
Changed line 3 from:
* [Recommended] Gary F. Marcus (2001). 'The Algebraic Mind : Integrating Connectionism and Cognitive Science' MIT Press. isbn:0262133792
to:
* [Recommended] Gary F. Marcus (2001). ''The Algebraic Mind : Integrating Connectionism and Cognitive Science'' MIT Press. isbn:0262133792