認知アーキテクチャACT-R

date_range 20/05/2021 10:53

ACT-R

• Anderson, J. R., Bothell, D., Byrne, M. D., Douglass, S., Lebiere, C., & Qin, Y . (2004). An integrated theory of the mind. Psychological Review 111, (4). 1036-1060.
• http://act-r.psy.cmu.edu/

ACT-R Assignment Unit8

date_range 16/12/2020 14:36

顔分類タスク

---

• a simplification of an experiment which was performed by Robert M. Nosofsky
• The experiment
  • trained on learning 10 faces (5 of each categories)
  • varied along:
    • eye height (EH)
    • eye seperation (ES)
    • nose length (NL)
    • mouth height (MH)
  • testing phase
    • new faces
    • old faces
• The model:
  • Not involve any learning mechanism

  • training information pre-encoded in declarative memory, model reset on every trial

  • 流れ：
    • presented with the attributes one at a time
      • name-value
    • collect those attributes into a single chunk (encoded in the imaginal buffer)
    • using the chunk to retrieve a best matching in the DM
    • based on the chunk, make a category choice for the current stimuli
  • Hint
    • first thing to do is to creating stimulus representation from the individual attributes
    • no more than 5 productions
    • use procedual partial matching
    • use dynamic pattern matching
• The Stimulus Attributes - Goal buffer in the initial state
  • CHUNK1-0
    • NAME EH
    • VALUE 0.7
    • STATE ADD-ATTRIBUTE
    • We have to convert the numeric number 0.7 into a symbolic description stored in a slot of the imaginal buffer
    • small
    • medium
    • large
    • use procedual partial matching and similarity hook to convert numeric attributes t label attributes

ACT-R Tutorial Unit8

date_range 15/12/2020 06:42

Advanced Production Techniques

---

• Talk about 2 additional mechanisms in the procedural system
  • procedural partial matching
  • dynamic pattern matching

ACT-R Assignment Unit7

date_range 06/12/2020 14:36

英語過去式を生成する認知モデル

• 

ACT-R Tutorial Unit7

date_range 02/12/2020 14:36

Production Rule Learning