an artificial mind via cognitive modular neural architecture pentti o. a. haikonen nokia research...
TRANSCRIPT
An Artificial Mind viaCognitive Modular Neural Architecture
Pentti O. A. Haikonen
NOKIA RESEARCH CENTER
P.O. Box 407
FIN-00045 NOKIA GROUP, Finland
© Pentti O A Haikonen / Nokia Research Center 2000
What is Mind?
An artificial mind should also have these
to qualify as a mind of any credibility.
Human mind is characterized by
Inner imagery, Inner speech, Sensations,Emotions
© Pentti O A Haikonen / Nokia Research Center 2000
-The human brain processes information with meaning and importance
-There is “an unified experience” the instantaneous sensory information from multiple sensors is bound together and is coupled to the system’s knowledge and emotional state so that a stream of interpretation, meaning and mental responses arises - the flow of inner imagery, inner speech, feelings
-This style of information processing -cognitive information processing- is completely different from present day computers
Human Mind
© Pentti O A Haikonen / Nokia Research Center 2000
What is Involved in Cognition?-Meaning and understanding -Perception and recognition-Prediction -Priming -Attention-Match/mismatch/novelty detection-Learning and memory-Judgement, good/bad-Pain and pleasure-Emotions-Motivation, needs, drives, goals-Deduction, reasoning, planning-Language-Consciousness?
© Pentti O A Haikonen / Nokia Research Center 2000
What is Involved in Consciousness?
-Awareness of environment
-Awareness of own body
-Awareness of qualia, how it feels
-Introspection of thoughts, emotions and feelings
-Awareness of past, present and expected future
-Awareness of self, one’s own existence
-Awareness and ability to report the existence of one’s inner imagery and speech as such
© Pentti O A Haikonen / Nokia Research Center 2000
Inner Imagery, Inner Speech
-The flow of inner speech, inner imagery is typical to human cognition
- Inner speech and inner imagery are also understood as verbal and visual thinking
-Inner speech, imagery, emotions and sensations are also the contents of our consciousness
© Pentti O A Haikonen / Nokia Research Center 2000
Steps towards Machine Mind
1. Devise suitable information representation method (distributed signal representation)
2. Devise an elementary processing unit for the above (non-numeric associative neuron)
3. Devise system architecture that can support inner imagery etc. and the cognitive processes (reentrant modular architecture)
© Pentti O A Haikonen / Nokia Research Center 2000
The Neuron
input signal output signal
associative signal array
m signals
match signal
mismatch signal
novelty signal
threshold
- Preservation of the input signal meaning- Correlative Hebbian (Associative) learning
- Resolves match/mismatch/novelty states
- Non-numeric
© Pentti O A Haikonen / Nokia Research Center 2000
input signal array
n signals
output signal array
n signals
associative signal array
m signals
match signal
mismatch signal
novelty signal
NEURON GROUP
THR
ES
HO
LD
Neuron Group as the Basic Signal Processor
- Compression or generalization when n < m
- Association of signal arrays to each other
- Associative evocation of output signal arrays
- Amplification or priming by the associative signal array
- Resolves match/mismatch/novelty between input and evocation
© Pentti O A Haikonen / Nokia Research Center 2000
The Reentrant Loop -Key to Inner Imagery
sensor&featuredetectors
representation
object
neuron
representation representationfeedbackneurons
inputsensed evoked
neuron
neuron
associative neuron system
(feature signals)
percept
percept
percept
output
-Perception with and without priming
-Reverberating short term memory
-Translation of output representations into percepts (inner imagery, inner speech), Introspection
-Grounding of meaning
-Percept - the “official” output to other modules
© Pentti O A Haikonen / Nokia Research Center 2000
preprocess
preprocess
reward
punish
m/mm/n
m/mm/n
neuronsfeedback
neuronsfeedback
m/mm
m/mm
displeasurepleasure/
system
WTA
WTA
M/MM
system
association neuron groups
association neuron groups
Associative
system
m/mmP/DP
system
Sensory module 1
Sensory module 2
sensoryinput
sensoryinput
percept
percept
The Cognitive System
The complete system consists of: -Multiple associatively cross-connected sensory modules-Pleasure/displeasure system-Match/mismatch/novelty detection
© Pentti O A Haikonen / Nokia Research Center 2000
preprocess
preprocessvisual
reward
attention &
punish
m/mm/n
m/mm/n
shape
size
color
shape
size
color
pospos
gazedirection
input
input
percept
percept
percept
word
neuronsfeedback
neuronsfeedback
neuronsfeedback
neuronsfeedback
m/mm
m/mm
text
displeasurepleasure/
system
percept
WTA
percept WTA
WTA
WTA
WTA
Linguistic system
Visual system
Visualattentionfocussystem
P/DP system
association neuron groups
association neuron groups
association neuron groups
association neuron groups
association neuron groups
M/MMsystem
Associative
system
m/mm
© Pentti O A Haikonen / Nokia Research Center 2000
The Simulation System
© Pentti O A Haikonen / Nokia Research Center 2000
neurons
system
m/mm/n
Winner-Take
s-All
pleasure/displeasure
input
D D
shape
size
color
name
pos
D
p/dp
feedback
x
x
x
x
x
x
x
m/mm
m/mm
m/mm
m/mm
m/mm
m/mm
predw(156,468)word(156)
primw1(468)
primw2(468)
primw3(468)
primw4(468)
primw5(468)
primw6(468)
primw7(468)
oword(156)
m
m
m
m
m
m
m
text
punish
reward
systemM/MM
M/MM
neurons
m/mm/n
shape
feedback
WTA
neurons
m/mm/n
feedback
WTA
neurons
m/mm/n
feedback
WTA
size
color
m/mm
m/mm
word/size
word/color
visual input&
preprocess
WTA
focus pos
focus pos change det.
stepper
m/mm/n
wnw1(156,40)
wnw2(156,3)
wnw3(156,3)
wnw4(156,46)
wnw5(156,5)
wnw6(156,2)
wnw7(156,2)
snw(3,156)
cnw(3,156)
featr(40)
siz(3)
colr(3)
pos(36) fpw(36,40)
wenw(2,156)
m/mmword/shape
focus pos/shape
ftrnw1(40,156)
ftrnw2(40,36)
m/mm
ftrpw(40,1)
ftrdpw(40,1)
p/shape
dp/shape
© Pentti O A Haikonen / Nokia Research Center 2000
time
The Simulation SystemNaming entities
1. Point the object with a laser pointer
2. Type in a name
3. Push “Emph” and “Enter”
© Pentti O A Haikonen / Nokia Research Center 2000
The Simulation SystemTeaching categories; category “shape”
© Pentti O A Haikonen / Nokia Research Center 2000
The Simulation SystemNaming an entity with shape, color and size attributes
© Pentti O A Haikonen / Nokia Research Center 2000
The Simulation SystemDeduction by evoked inner imagery, answering a question
The word “square” has not been explicitly associated to “dollar”!
© Pentti O A Haikonen / Nokia Research Center 2000
The Simulation SystemDeduction by evoked inner imagery, answering a question
The word “green” has not been explicitly associated to “dollar”! © Pentti O A Haikonen / Nokia Research Center 2000
Deduction by evoked inner imagery, contradiction detection
The Simulation System
© Pentti O A Haikonen / Nokia Research Center 2000
Deduction by evoked inner imagery, affirmation detection
The Simulation System
© Pentti O A Haikonen / Nokia Research Center 2000
Emotional significance; detection of an emotionally significant entity from noise
The Simulation System
© Pentti O A Haikonen / Nokia Research Center 2000
Visual search of a given entity; the search is completed when a sensed object matches the inner image of the object to be searched. No pattern matching is done however!
The Simulation System
© Pentti O A Haikonen / Nokia Research Center 2000
The Simulation System
-Verbal sequences, reproduction
-Sequences as serial associative prediction; detection of mismatch between prediction and actual percept
© Pentti O A Haikonen / Nokia Research Center 2000
Conclusions 1
A modular non-numeric neural network has been devised that
-Operates with inner imagery, inner speech-Acquires information via perception-Acquires information about its inner states via introspective perception-Is able to learn and generalize (and fast!)-Has cognitive functions similar to human brain
What remains to be demonstrated:-Actual motor output systems-The effect of needs, drives, planning, will-Personal history, sense of time-Self concepts
© Pentti O A Haikonen / Nokia Research Center 2000
Conclusions 2
TOWARDS CONSCIOUS MACHINES?
This system has the flow of inner imagery and inner speech, the hallmarks of human consciousness
However, the system is not yet able to report on its own that it exists, that it has inner imagery and inner speech
A system’s ability to report on its own that it has inner speech, produced by the system self, could be used as a test for machine self-consciousness
The author would like to see the Turing test be replaced by this one.
© Pentti O A Haikonen / Nokia Research Center 2000
Dr. Pentti O. A. Haikonenprincipal scientist
Cognitive Technology
NOKIA RESEARCH CENTER
P.O. Box 407
FIN-00045 NOKIA GROUP Finland
preprocess
preprocessvisual
reward
attention &
punish
m/mm/n
m/mm/n
shape
size
color
shape
size
color
pospos
gazedirection
input
input
percept
percept
percept
word
neuronsfeedback
neuronsfeedback
neuronsfeedback
neuronsfeedback
m/mm
m/mm
text
displeasurepleasure/
system
percept
WTA
percept WTA
WTA
WTA
WTA
Linguistic system
Visual system
Visualattentionfocussystem
P/DP system
association neuron groups
association neuron groups
association neuron groups
association neuron groups
association neuron groups
M/MMsystem
Associative
system
m/mm
Thank You !An Artificial Mind via
Cognitive Modular Neural Architecture
© Pentti O A Haikonen / Nokia Research Center 2000