Kognitive Architekturen

1
ACT-R 6.0 (Cognitive Science Prosem) Wayne D.
Gray Rensselaer Polytechnic Institute
CogWorks Laboratory grayw_at_rpi.edu Mike
Schoelles Rensselaer Polytechnic Institute
CogWorks Laboratory schoem_at_rpi.edu
2
Tutorial Overview

• Cognitive Architecture/Modeling Overview
• ACT-R Theory - Symbolic level
• Addition,counting and letter models
• ACT-R Theory Sub-symbolic level
• Sternberg and Building Sticks models
• Production Compilation

3
What is a Cognitive Architecture?

• Infrastructure for an intelligent system
• Cognitive functions that are constant over time
  and across different task domains
• Analogous to a building, car, or computer

4
Unified Theories of Cognition

• Account of intelligent behavior at the
  system-level
• Newells claim
• You cant play 20 questions with nature and win

5
Integrated Cognitive Architecture

• Cognition does not function in isolation
• Interaction with perception, motor, auditory,
  etc. systems
• Embodied cognition
• Represents a shift from
• mind as an abstract information processing
  system
• Perceptual and motor are merely input and output
  systems
• Must consider the role of the environment
• Other body processes
• Effects of caffeine, stress and other moderators

6
Motivations for a Cognitive Architecture

• 1. Philosophy Provide a unified understanding
  of the mind.
• 2. Psychology Account for experimental data.
• 3. Education Provide cognitive models for
  intelligent tutoring systems and other learning
  environments.
• 4. Human Computer Interaction Evaluate artifacts
  and help in their design.
• 5. Computer Generated Forces Provide cognitive
  agents to inhabit training environments and
  games.
• 6. Neuroscience Provide a framework for
  interpreting data from brain imaging.
• 7. All of the above

7
Requirements for Cognitive Architectures

• 1. Integration, not just of different aspects of
  higher level cognition but of cognition,
  perception, and action.
• 2. Systems that run in real time.
• 3. Robust behavior in the face of error, the
  unexpected, and the unknown.
• 4. Parameter-free predictions of behavior.
• 5. Learning.

8
Newells Time Scale of Human Activity (amended)
9
Taxonomy
Computational Cognitive Models
Connectionist
Home grown -- one-off code
Symbolic
Cognitive Architectures
Other AI
other
Production System
Hybrid
Symbolic only
ACT-R 6.0
SOAR
EPIC
10
(No Transcript)
11
History of the ACT-framework
Predecessor HAM (Anderson Bower
1973) Theory versions ACT-E (Anderson,
1976) ACT (Anderson, 1978) ACT-R (And
erson, 1993) ACT-R 4.0 (Anderson Lebiere,
1998) ACT-R 5.0 (Anderson Lebiere,
2001) Implementations GRAPES (Sauers
Farrell, 1982) PUPS (Anderson Thompson,
1989) ACT-R 2.0 (Lebiere Kushmerick,
1993) ACT-R 3.0 (Lebiere, 1995) ACT-R
4.0 (Lebiere, 1998) ACT-R/PM (Byrne,
1998) ACT-R 5.0 (Lebiere, 2001) Windows
Environment (Bothell, 2001) Macintosh
Environment (Fincham, 2001) ACT-R
6.0 (Bothell, 2004??)
12
Other Cognitive Architectures

• Soar
• Production rule system
• Organized in terms of operators associated with
  problem spaces
• Goal oriented
• Sub-goaling
• Learning mechanism - Chunking
• EPIC -- Executive Processes in Cognition
• Parallel firing of production rules (in
  principle)
• Well developed visual and motor system
• Emphasis on executive processes

13
MetaIssues

• The divide between high-level architectures and
  low-level (primarily connectionist ones) is
  mainly a levels issue
• Modeling in high-level architectures reflects a
  concern with the task structure of behavior and
  can be considered a form of task analysis

14
MetaIssues

• The problems tackled by the two sets of
  approaches tends to reflect this difference
• From this perspective ACT-R is a major success in
  re-use -- low level parameters are reused in
  most ACT-R models.
• The higher level production rules differ in part
  because they reflect the task analysis for the
  different tasks being modeled

15
MetaIssues

• From this perspective the focus on reusability
  should focus on
• Low-level productions that control the
  interleaving of cognitive, perceptual, and action
  at the 1/3 sec level of analysis
• Not on
• High-level productions that implement
  task-specific executive control

16
MetaIssues

• Low-Level Productions Implement Interactive
  Routines

17
MetaIssues

• Interactive Routines
• Neurocognitive evidence increasing stresses the
  modular nature of human cognition
• But -- these modules are constrained by their
  need to work together to survive in the world
• Interactive routines can be viewed as the basic
  level elements of embodied cognition
• Provide constraints on the input/output functions
  for perception, attention, memory, and motor
  systems
• Notion is that the use of cognitive, perceptual,
  and motor resources is optimized via the
  selection of one set of interactive routines over
  another

18
ACT-R Overview

• Modules (buffers)
• Knowledge Representation
• Symbolic/Sub-symbolic
• Performance/Learning

19
ACT-R Applications 559 Papers Listed on
http//act-r.psy.cmu.edu/publications/index.php In
the following areas as of 2006-01-24
20
Interactive Session

• Load and run Addition model

21
Addition model exercise

• In this exercise you will load a simple model and
  run it to see how a model runs. You will also
  get some experience with the interface.
• 1. Open model
• Click on the "Open Model" button on the
  Environment Control Pane, and select the Addition
  model. This will open up the model so that you
  can see it and its parts.
• You should be able to see the working memory
  elements in the model (window "Chunk"), the
  productions (Production window). There are three
  further windows, Chunk Type, Command, and
  Miscellaneous, that we will cover later.
• You should briefly examine the chunk and
  production contents. You may note that there
  about 11 pieces of working memory, and just 4
  rules in this system.

22

• 2. Run the model
• You can run the model using the Lisp command
  line, but we will use the environment because it
  provides a recognition-based interface rather
  than a recall-based interface.
• You should first click on "Reset" this will
  reset the model and make it ready to run. You
  can do this to a model that has run as well, or
  has been stopped in the middle of a run.
• You can run the model by clicking on the "Run"
  button. A trace of the model will appear in the
  (Lisp) "Listener" window. You can see how the
  order that rules are selected and fired, as well
  as when chunks are retrieved from memory by the
  rules.

23

• 3. Inspect the model
• Click on "Declarative viewer" in the Control Pane
  to bring up an inspector window for the
  declarative memory elements. If you scroll, you
  can find the chunks a-j and second-goal. Pay
  most attention to their structure, and note that
  they have several parameters. These parameters
  are used to compute how fast they are used and if
  they can be retrieved. With learning and use,
  the activation, for example, goes up. These are
  covered later in this tutorial.
• The Procedural viewer provides a view onto the
  rules.

24
End of First Tuesday

• Homework
• Anderson, J. R., Bothell, D., Byrne, M. D.,
  Douglas, S., Lebiere, C., Quin, Y. (2004). An
  integrated theory of the mind. Psychological
  Review, 111(4), 1036-1060.
• Change addition model to subtraction model

25
ACT-R 6.0 Architecture
26
ACT-R 6.0 Mapping to the Brain
Intentional Module (not identified)
Declarative Module (Temporal/Hippocampus)
Retrieval Buffer (VLPFC)
Goal Buffer (DLPFC)
Matching (Striatum)
Productions (Basal Ganglia)
Selection (Pallidum)
Execution (Thalamus)
Visual Buffer (Parietal)
Manual Buffer (Motor)
Manual Module (Motor/Cerebellum)
Visual Module (Occipital/etc)
Environment
27
ACT-R Assumption Space
28
ACT-R Knowledge Representation
? goal buffer ? visual buffer ? retrieval buffer
29
Declarative Memory Syntax
Chunk type ((chunk-type type-name slot-name1
(slot-namek init-val)..slot-namen)
(chunk-type count-order first second) (chunk-type
add arg1 arg2 sum count)
Chunk Instantiation (add-dm (chunk-name isa
type-name slot-name slot-value) )
(add-dm (a ISA count-order first 0 second 1)
(b ISA count-order first 1 second 2)
(second-goal ISA add arg1 5 arg2 2) ) sum and
count are nil
30
Declarative Memory

• Chunks that are added explicitly
• Add-dm
• Chunks merge into DM from buffers
• All buffers chunks go to DM when cleared
• Mergings are the references for BLL
• Not the LHS usage as in ACT-R 5
• Because buffers hold copies, DM chunks cant be
  changed from within a production
• Previously it was a recommendation

31
Addition Fact Example
(Chunk-type addition-fact addend1 addend2 sum)
32
Addition Example
(CLEAR-ALL) (DEFINE-MODEL addition) (CHUNK-TYPE
addition-fact addend1 addend2 sum) (CHUNK-TYPE
integer value) (ADD-DM (fact34 isa
addition-fact addend1 three addend2 four
sum seven) (three isa integer value 3)
(four isa integer value 4) (seven isa
integer value 7)
33
Addition Fact Example
ADDITION-FACT
3
7
VALUE
isa
VALUE
ADDEND1
SUM
FACT34
THREE
SEVEN
ADDEND2
4
isa
isa
FOUR
VALUE
isa
INTEGER
34
A Production is 1. The greatest idea in
cognitive science. 2. The least appreciated
construct in cognitive science. 3. A 50
millisecond step of cognition. 4. The source of
the serial bottleneck in otherwise parallel
system. 5. A condition-action data structure
with variables. 6. A formal specification of
the flow of information from cortex to basal
ganglia and back again.
35
Productions
modularity abstraction goal/buffer
factoring conditional asymmetry
Key Properties
Structure of productions
(
p
name
Specification of Buffer Tests
condition part
gt
delimiter
Specification of Buffer Transformations
action part
)
36
Productions LHS

• Only four possible conditions available
• buffergt
• Test the chunk in the buffer just like in 5
• !eval! or !safe-eval!
• !bind! or !safe-bind!
• Same as in ACT-R 5
• Safe- versions accepted by production compilation
• ?buffergt
• Query the buffer or its module
• Come back to queries later

37
Possible RHS actions

• buffergt
• -buffergt
• buffergt
• !eval! and !safe-eval!
• !bind! and !safe-bind!
• !output!
• !stop!

38
RHS actions

• buffergt
• !eval! and !safe-eval!
• !bind! and !safe-bind!
• !output!
• All the same as in ACT-R 5
• The safe- versions do not inhibit the production
  compilation mechanism
• !stop!
• Not actually new, but does work now
• Generates a break event in the scheduler
• Terminates the current run command

39
RHS buffergt

• buffergt isa chunk-type
• modifier slot request parameter value
• or
• buffergt chunk-reference
• Sends a request to the module
• Always clears the buffer implicitly
• Essentially the same as ACT-R 5

40
Buffer queries

• Replaces the -state buffers
• Syntax
• ?buffergt
• - query value
• Either true or false
• No bindings
• Must all be true for production to match
• Examples
• ?retrievalgt ?visualgt
• state busy - state error
• buffer empty buffer check

41
Queries continued

• Every buffer/module must respond to
• State
• Values busy, free, or error
• Buffer
• Values full, empty, requested or unrequested
• Others can be added by a module writer
• Modality for the current PM modules for example

42
Production Syntax
(P initialize-addition goalgt ISA
add arg1 num1 arg2
num2 sum nil gt goalgt
sum num1 count 0
retrievalgt isa count-order
first num1 ?retrievalgt state
free )
(P increment-sum goalgt ISA add
sum sum count count
retrievalgt ISA count-order
first sum second newsum gt
goalgt sum newsum retrievalgt
isa count-order first
count )
43
(p got-number goalgt isa make-a-call
who person retrievalgt isa phone-numb
er who person where office
ph-num num gt !output! (The phone
number for person is num) goalgt
isa dial-number who person
ph-num num current-digit 1 )
(p dial-a-digit goalgt
isa dial-number ph-num num
current-digit digit state nil gt
!bind! d (get-digit digit num)
visual-locationgt isa visual-location
value d goalgt state dialing )
44
(P increment goalgt ISA
count-from start num1 - end
num1 step counting
retrievalgt ISA count-order
first num1 second num2 gt
goalgt start num2 retrievalgt
ISA count-order first
num2 !output! (num1) )
(p read-choose goalgt isa read-letters
state verify-choose visualgt isa text
value "choose" gt goalgt state
find-letter )
45
(P rotate-counter-clockwise goalgt ISA
translaterotate step
counter-clockwise reference-y y axis
axis form form !bind! y1
( y 15)) retrievalgt ISA
point-i gt screen-y y1 no longer can do
(!eval! ( y 15)) gt !eval!
(rotate-counter-clockwise form axis) )
(P get-direction goalgt ISA
make-report step find-point
retrievalgt ISA point-i
screen-x x screen-y y gt
goalgt step find-direction !bind!
x1 ( x 15) retrievalgt ISA
direction lt screen-x x1 no longer
(!eval! ( x 15)) )
46
Interactive Session

• Load and run Counting model

47
Count model

• This model works much like the previous model,
  but prints out its count.
• 1. Open the model
• Either quit and restart your Lisp, or else click
  on "Close Model".
• Open the Count model by clicking on "Open Model"
  and then selecting the Count model.
• Run the model to see its trace, and examine its
  rules and chunks.
• 2. Using the Stepper
• Click on "Stepper", and a stepper window should
  appear.
• Reset the model, and then click on the run
  button. This starts the stepper. You can now
  step through the model by clicking on the "Step"
  button on the Stepper.
• As you step through the model, you should be able
  to see most of the mechanisms in ACT-R now, the
  productions, how they are matched, the chunks,
  and how they are retrieved, and the buffers
  (click on Buffer Viewer to see the buffers and
  their contents).

48

• 3. Checking on a rule that does not fire.
• After you have run the model a few steps, click
  on the Procedural Viewer. Select a rule in the
  dialogue box, and see why it does not fire.
• 4. Edit the model
• Look at the model and consider how to have it
  count backwards.
• You can change the production rules in the
  Production window. After you make changes, save
  the model (it will automatically increment).
  Close the model and reopen it to try your new
  model.

49
The Modules(reprise)

• Cognition
• Memory
• Vision
• Motor
• Audition
• Speech

50
ACT-R 6.0 Buffers 1. Goal Buffer (goal,
goal) -represents where one is in the
task -preserves information across production
cycles
2. Retrieval Buffer (retrieval,
retrieval) -holds information retrieval from
declarative memory -seat of activation
computations 3. Visual Buffers -location
(visual-location, visual-location) -visual
objects (visual, visual) -attention switch
corresponds to buffer transformation 4. Auditory
Buffers (aural, aural) -analogous to visual 5.
Manual Buffers (manual, manual) -elaborate
theory of manual movement include feature
preparation, Fitts law, and device properties 6.
Vocal Buffers (vocal, vocal) -analogous to
manual buffers but less well developed
51
Cognition

• Executive Control - Production System
• Serial
• Parallel at sub-symbolic level
• Utility selects production to fire
• Utility benefit - cost
• Benefit probability of success value of
  achieving goal

52
Production System Cycle

• Match conditions of all rules to buffers
• Those that match enter the conflict set
• Conflict resolution selects a rule to fire
• Action side of rule initiates changes to one or
  more buffers
• If no production can match and no action is in
  progress then quit else repeat

53
Goal directed

• Represents what you are trying to do
• A declarative memory element that is the focus of
  internal attention
• (goal-focus second-goal)

54
Memory Module

• Activation based
• Frequency and recency
• Contextual cues
• Cognition
• Requests retrieval
• Specifies constraints
• Partial matching
• Memory
• Parallel search of memory to match constraints
• Calculates activation of matching chunks
• Returns most active chunk

55
Vision Module

• ACT-Rs eyes
• Dorsal where system
• Ventral what system

56
Where System

• Cognition
• Requests pre-attentive visual search
• Specifies a set of constraints
• Attribute/value pairs
• Properties or spatial location
• e.g. color red, screen-x greater-than 150
• Where system
• Returns a location chunk
• Specifies location of an object whose features
  satisfy the constraints
• Onsets
• Features are held in vision modules memory

57
Visual location
(chunk-type visual-location screen-x screen-y
distance kind color value size
nearest)
(P find-unattended-letter goalgt ISA
read-letters state start gt
visual-locationgt ISA
visual-location attended nil goalgt
state find-location )
(P attend-letter goalgt ISA
read-letters state find-location
visual-locationgt ISA
visual-location ?visualgt state
free gt visualgt ISA
move-attention screen-pos
visual-location goalgt state
attend )
58
What System

• Cognition
• Requests move attention
• Provides location chunk
• What System
• Shifts visual attention to that location
• Encodes object at that location
• Added to Declarative Memory
• Episodic representation of visual scene
• Places encoding in visual buffer
• Calculates latency
• EMMA

59
Visual Object
(chunk-type visual-object screen-pos value
status color height width)
(P encode-letter goalgt ISA
read-letters state attend visualgt
ISA text value
letter gt goalgt letter letter
state respond )
(P attend-letter goalgt ISA
read-letters state find-location
visual-locationgt ISA
visual-location ?visualgt state
free gt visualgt ISA
move-attention screen-pos
visual-location goalgt state
attend )
60
Visual State

(P attend-letter goalgt ISA
read-letters state find-location
visual-locationgt ISA
visual-location ?visualgt state free
gt visualgt ISA
move-attention screen-pos
visual-location goalgt state
attend )
61
Vision Module Memory
62
Motor Module

• ACT-Rs Hands
• Based on EPICs Manual Motor Processor
• Movement Styles
• Phased Processing

63
Motor Syntax
(P respond goalgt ISA
read-letters letter letter
state respond ?manualgt
state free gt manualgt ISA
press-key key letter goalgt
state stop )
64
Movement Styles
Style
Punch (hand finger)
HFRT(hand finger r theta)
Peck
Peck-recoil
Ply (hand r theta)
Hand
Cursor
65
Movement Styles

• Ply - moves a device (e.g. move-mouse) to a given
  location
• Punch - pressing a key below finger or
  click-mouse
• Peck - directed movement of finger to a location
  followed by keystroke
• Peck-recoil - same as peck but finger moves back

66
Phased Processing (1)

• Preparation Phase
• Hierarchical feature preparation
• Style-gthand-gtfinger
• Prep time depends on
• Complexity of movement
• Number of features
• State buffer set to prep busy

67
Phased Processing (2)

• Initiation (fixed 50 ms)
• Execution
• Time depends on
• Type of movement
• Minimum execution time
• Distance
• Fitts Law
• Allow overlapping of preparation and execution

68
Audition Module

• Simulated perception of audio
• Memory of features
• Temporal-extent - sound events
• Tones, digits, and speech
• Attributes
• Onset, duration, delay, recode time

69
Audition Module Syntax
(chunk-type audio-event onset offset pitch kind
location) (chunk-type sound kind content event)
(defp alpha-task/listen goalgt isa alpha-task
step listen gt aural-locationgt
isa audio-event onset highest
attended NIL goalgt step check-feature
)
(defp alpha-task/check-feature
goalgt isa alpha-task step check-feature
aural-locationgt isa audio-event
?auralgt state free gt
auralgt isa sound event aural-location
goalgt step attend-sound)
70
Audition Module Processing

• Parallels vision system
• Cognition
• Specifies a set of constraints
• Attribute/value pairs
• Audition
• Returns a location chunk
• Cognition
• Requests shift of auditory attention providing
  the locationchunk
• Audition
• Encodes the sound

71
Device Interface

• Simulated device with which ACT-R interacts
• Contains graphical objects
• Typically a Window
• Can be entire screen
• Interaction
• Constructing vision systems iconic memory (sets
  of features) from graphical objects
• Handle mouse and keyboard actions

72
(No Transcript)
73
(No Transcript)
74
Handy commands

• (dm)
• (sdm slot value)
• (sdp chunk-name)
• (get-chunk chunk-name) returns chunk structure
• (get-chunk-type 'name) gets type structure from
  type name
• (get-module module-name)
• (my-name (get-module vision))
• (print-module-state (get-module vision))
• (current-mp)
• (current-device)
• (current-device-interface)
• (buffers)
• (buffer-chunk buffer-name)
• (buffer-read 'buffer-name)
• (chunk-slot-value chunk-name slot-name)
• (pprint-a-chunk chunk-name)
• (sdp-fct (no-output (dm)))
• (sdp-fct (no-output (sdm isa ...)))
• (gethash vision (act-r-modules-table
  modules-lookup))

75
Interactive Session

• Load and run Letter model

76
Sub-symbolic level

• Sub-symbolic learning allow the system to adapt
  to the statistical structure of the environment
• Production Utilities are responsible for
  determining which productions get selected when
  there is a conflict.
• Chunk Activations are responsible for determining
  which (if any chunks) get retrieved and how long
  it takes to retrieve them.
• Chunk Activations have been simplified in ACT-R
  5.0 and a major step has been taken towards the
  goal of parameter-free predictions by fixing a
  number of the parameters.

77
Parameters

• Noise
• Utility and activation
• Learning
• Activation - frequency and recency
• Utility - probability and cost
• Thresholds
• Utility and activation

78
Sub-symbolic ACT-R theory

• Activation equation
• Production Utility equation
• Production Compilation

79
Activation
Seven
Sum
Addend1
Addend2
Chunk i
Three
Four
B
i
S
ji
Goalgt
Retrievalgt

isa
write



isa
addition-fact
relation sum
Conditions
Actions
addend1 Three
arg1 Three
Sim
addend2 Four
arg2 Four
kl
80
Chunk Activation

• Activation Base Level Associative
  Partial Matching Noise
• Reflects the degree that a chunk will be useful
  in the current context based upon past
  experiences
• Base level - general past usefulness
• Associative - relevance to current context
• Partial Matching - relevance to the specific
  match
• Noise - stochastic, non-deterministic behavior

81
Base-level Activation
Ai Bi
The base level activation Bi of chunk Ci reflects
a context-independent estimation of how likely Ci
is to match a production, i.e. Bi is an estimate
of the log odds that Ci will be used. Two
factors determine Bi frequency of using
Ci recency with which Ci was used Bi is set
globally using the set-all-base-levels
function Bi is set for individual chunks using
(set-base-level (get-wme 'set-dow) '(100 -100))
if learning on (set-base-level (get-wme
'set-dow) 50.0) if learning off
82
Base-level Learning

• Base-Level Activation reflects the log-odds that
  a chunk will be needed.
• The odds that a fact will be needed decays as a
  power function of how long it has been since it
  has been used.
• The effects of multiple uses sum in determining
  the odds of being used.

83
Base-level Learning Equations
84
Chunk Presentation

• Creation
• Initialization (add-dm)
• Encode from environment (visual)
• New goal (goalgt isa )
• Merging
• When goal is cleared (-goal or goal)
• Merged if matches a current chunk
• Harvest of a retrieval
• retrieval
• Could get multiple presentations from one
  retrieval

85
Source Activation
? Wj Sji
j
The source activations Wj reflect the amount of
attention given to elements of the current goal.
ACT-R assumes a fixed capacity for source
activation
W ? Wj reflects an individual difference
parameter.
86
Associative Strengths
? Wj Sji
The association strength Sji between chunks Cj
and Ci is a measure of how often Ci was needed
(retrieved) when Cj was element of the goal, i.e.
Sji estimates the log likelihood ratio of Cj
being a source of activation if Ci was retrieved.
Sji S - ln(FANj) FANj of chunks in which
chunk j is the value of a slot 1 for chunk j
being associated with itself S maximum
associative strength- a constant (mas)
87
(No Transcript)
88
Partial Matching

• The mismatch penalty is a measure of the amount
  of control over memory retrieval MP 0 is free
  association MP very large means perfect
  matching intermediate values allow some
  mismatching in search of a memory match.
• Similarity values between desired value k
  specified by the production and actual value l
  present in the retrieved chunk. This provides
  generalization properties similar to those in
  neural networks the similarity value is
  essentially equivalent to the dot-product between
  distributed representations.

89
Noise

• Generated according to a logistic distribution
  characterized by parameter s. The mean the
  distribution is 0 and variance ??.

• Noise provides the essential stochasticity of
  human behavior
• Noise also provides a powerful way of exploring
  the world
• Activation noise is composed of two noises
• A permanent noise accounting for encoding
  variability
• A transient noise for moment-to-moment variation

90
Latency

• Retrieval time for a chunk is a negative
  exponential function of its activation
• A activation of chunk being retrieved
• F latency scale factor (set globally with the
  lf parameter)
• If no chunk matches or no chunk is above
  retrieval threshold

t retrieval threshold
91
Probability of Retrieval

• Probability of retrieval of a chunk follows the
  Boltzmann (softmax) distribution

• The chunk with the highest activation is
  retrieved provided that it reaches the retrieval
  threshold ?
• For purposes of latency and probability, the
  threshold can be considered as a virtual chunk

92
Interactive Session

• Load and run Sternberg model

93
Production Utility
P is expected probability of success G is value
of goal C is expected cost Noise
??????????????s? where s is set globally by the
egs parameter
t reflects noise in evaluation and is like
temperature in the Bolztman equation
94
P C
95
Interactive Session

• Load and run Building Sticks model

96
Production Compilation The Basic Idea
(p read-stimulus goalgt isa goal
step attending state test visualgt
isa text value val gt retrievalgt
isa goal relation associate arg1
val arg2 ans goalgt relation
associate arg1 val step testing) (p
recall goalgt isa goal relation
associate arg1 val step testing
retrievalgt isa goal
relation associate arg1 val arg2
ans gt manualgt isa
press-key key ans goalgt
step waiting) (p recall-vanilla goalgt
isa goal step attending state test
visualgt isa text value "vanilla gt
manualgt isa press-key
key "7" goalgt relation
associate arg1 "vanilla" step
waiting)
97
Production Compilation The Principles
1. Perceptual-Motor Buffers Avoid compositions
that will result in jamming when one tries to
build two operations on the same buffer into the
same production. 2. Retrieval Buffer Except for
failure tests proceduralize out and build more
specific productions. 3. Goal Buffers Complex
Rules describing merging. 4. Safe Productions
Production will not produce any result that the
original productions did not produce. 5.
Parameter Setting Successes
Pinitial-experience Failures (1-P)
initial-experience Efforts (Successes
Efforts)(C cost-penalty)
98
References
Introduction to ACT 5.0 Tutorial by Christian
Lebiere, http//act-r.psy.cmu.edu/tutorials/ ACTR
5.0 Equations, Variables and Parameters by Jerry
Ball
99
Base Level Activation - Bi

• The base level activation Bi of a chunk Ci
  reflects a context-independent estimation of how
  likely Ci is to match a production, i.e. Bi is an
  estimate of the log odds that Ci will be used.
• (set-base-level (get-wme 'set-dow) '(100 -100))
  if learning on
• (set-base-level (get-wme 'set-dow) 50.0) if
  learning off

100
Associative LearningNot ACT-R 6.0