Cognitive architectures
Download as PPTX, PDF1 like639 views
The document discusses cognitive modeling, highlighting various approaches such as symbolic, sub-symbolic, and hybrid models. It explores distinctions between declarative and procedural memory, and different learning methods including Hebbian and supervised learning. Examples of programming environments and tools used for creating cognitive models are also provided.
![Production systems
For example (assuming words beginning
with capital letters are variables, and other
words are constants):
if [Person, age, Number] &
[Person, employment, none] &
[Number, greater_than, 18] &
[Number, less_than, 65]
then [Person, can_claim,
unemployment_benefit].
6](https://image.slidesharecdn.com/2015-03-10-cognitive-architectures-150112150923-conversion-gate01/85/Cognitive-architectures-6-320.jpg)
Cognitive architectures
- 1. CGSC 1001 Mysteries of the Mind by Jim Davies jim@jimdavies.org 1
- 2. Across fields, it’s a representation of something that excludes unimportant detail and information • A scale model of a home made of cardboard • A categorization scheme for the students in my class • A simulation of a hurricane 2
- 3. Typically, it’s a computer program that models some aspect of thought. For example, it might model how people do categorization, or how a mouse learns to navigate a maze The model makes predictions that can then be compared to data If the predictions match the data, it supports the theory underlying the model 3
- 4. A programming environment or set of tools for making cognitive models Typically, it includes constraints on how cognition works in all people– speed of learning, memory retrieval, etc., ignoring cultural and other learned aspects. Making a model in an architecture is easier in some ways and harder in others. 4
- 5. Symbolic • Operates at the level of discrete symbols Sub-symbolic • Operates using number representations, which in aggregate constitute symbols (we’ll have examples later Hybrid symbolic/subsymbolic Brain • Models cognition at the level of biology, but speak to cognitive issues 5
- 6. Production systems For example (assuming words beginning with capital letters are variables, and other words are constants): if [Person, age, Number] & [Person, employment, none] & [Number, greater_than, 18] & [Number, less_than, 65] then [Person, can_claim, unemployment_benefit]. 6
- 7. Declarative/procedural memory distinction • Justifications: HM and other brain damaged patients • Our inability to consciously retrieve and reflect on procedural memories. Goals are subset of declarative memory Production compilation models automatization 7
- 8. 8 Uses Hebbian learning to learn patterns. When it gets incomplete input, It can complete it based on the weights Between the nodes. It is unsupervised learning.
- 9. 9
- 10. 10 Supervised Learning Uses backpropogation Great for classification Needs thousands of Examples to work, Unlike human beings
Editor's Notes
- #2: Pictured: Catrinas, traditions figures of day of the dead celebrations in Mexico. From wikimedia commons.