1. Towards the New Era of
Artificial Cognitive Systems
Prof. Taymor M. Nazmy
Vice dean Faculty of computer science,
Ain Shams University, Cairo, Egypt
ntaymoor19600@gmail.com
2. Who is interested in the cognitive systems?
What is the future of cognitive systems?
What are the main cognitive processes?
What is the role of tacit knowledge in cognition
processes?
What are the progresses towards building a
cognitive systems?
3. How are the architectures of cognitive
systems looks like?
How can spike neural networks be used
to build an efficient perception for
cognitive systems?
What about artificial consciousness
systems?
6. 2002 2010 2020 2030
MissionComplexity
Biological Mimicking
Embryonics
Extremophiles
DNA
Computing
Brain-like
computing
Self Assembled Array
Artificial nanopore
high resolution
Mars in situ
life detector
Sensor Web
Biological nanopore
low resolution
Skin and Bone
Self healing structure
and thermal protection
systems
Biologically inspired
aero-space systems
Space Transportation
Autonomous, and thinking spacecraft
7. American inventor and futurist Raymond
Kurzweil
Predicts machines will be more intelligent
than humans in the near future
The Singularity is Near (2045)
2020s: Nanobot use in medical field
2029: First computer passes Turning test
2045: Singularity
8. ``From the discussion between Turing
and one of his colleagues (M. H. A.
Newman, professor of mathematics at
the Manchester University):
Newman: I should like to be there when
your match between a man and a
machine takes place,
Turing: Oh yes, at least 100 years, I
should say .
Turing paper was published in 1950.
9. From Ray Kurzwail, The Singularity Summit at Stanford
Future of computing
3d circuits
Quantum
computing
Molecular
electronics
Optical computing
DNA computing
New materials
10. Mr. Kelly and Mr. Hamm point out that
cognitive systems represent the third era in
the history of computing.
the first era, computers were essentially
tabulating machines. Next came the
programmable computing era which emerged
in the 1940s
11. The US report
recommends that the
U.S. designate as a
national priority
research and
development in
emerging technologies
that enhance human
abilities and
efficiencies by
combining
four major
"NBIC“
12. The US Government report launching
a Human Cognome Project,
comparable to the successful Human
Genome Project, to chart the structure
and functions of the human mind.
The ability to map the human brain
from the smallest cells to the large
structure of the brain would allow
scientists to expand into new
technological frontiers.
Human Cognome Project
13. For Build the scientific &
engineering foundations of
cognitive systems
Towards enhancing the future of
cognitive systems
15. New approaches towards understanding and solving key
issues related to the engineering of artificial cognitive systems
New approaches towards endowing robots with advanced
perception and action capabilities
New ways of designing and implementing complete robotic
systems to build a strong basis for research on ways of
reaching the long term goals inherent in this challenge and to
provide the means for carrying out that research;
New, scientifically grounded system architectures integrating
communication, control and cognitive capabilities
A framework to facilitate cross-fertilisation between academic
and industrial research efforts in robotics
16.
17. • In 1991-1992, the Philosophy Department merged with the Psychology
Department to form the Department of Philosophy, Psychology, and
Cognitive Science.
• In 2003, this became the Department of Cognitive Science, one of only
about 15 dedicated Departments of Cognitive Science in the world
• In 2004, a PhD program was created in Cognitive Science. In the Spring
of 2010, the B.S. undergraduate program in Cognitive Science was
approved in bout 20 faculty
• Relevant Laboratories / Research Groups:
• CogWorks Lab (Cognitive Modeling)
• RAIR Lab (Artificial Intelligence and Reasoning)
• PandA Lab (Perception and Action, Virtual Reality)
• Human-Level Intelligence Lab
• Cognitive Architecture Lab
• Cognitive Robotics Lab
Rensselaer Polytechnics Inst.
18. Cognitive Systems Engineering became one
of the major emerging themes within the
Engineering Department at the University of
Cambridge.
Cognitive Systems Engineering seeks to build
systems that monitor, support, aid, and extend
human cognitive processes
19. Neuromorphic engineering is a new
interdisciplinary subject that takes
inspiration from biology, physics,
mathematics, computer science and
electronic engineering to design artificial
neural systems, such as vision systems,
head-eye systems, auditory processors,
and autonomous robots, whose physical
architecture and design principles are
based on those of biological nervous19
20. The logical rule based principles of
programming that have made computers so
powerful are actually preventing them from
attaining human-like cognitive powers,
preventing them from becoming true thinking
machine.
Also, Artificial intelligence, fuzzy logic, and
neural networks have all experienced some
degrees of success, but machines still cannot
recognize pictures or understand language
22. Cognitive systems will require innovation
breakthroughs at every layer of information
technology, starting with nanotechnology and
progressing through computing systems
design, information management,
programming and machine learning, and,
finally, the interfaces between machines and
humans.
Advances on this scale will require
remarkable efforts and collaboration, calling
forth the best minds—and the combined
resources–of academia, government and
23. There are many key words we may found relate
cognitive science with the other sciences such as:
Cognitive linguistics, Augmented cognition,
Cognitive informatics, Cognitive robotics,
artificial Cognitive systems, Cognitive
radio,
cognitive computing and modeling,
Metacognition, Cognitive ergonomics or,
artificial life, adaptive behavior,
computational neuroethology.
24. Cognitive computing such as:
concepts, discover relations and rules,
sequences, actions, perceptions,
accumulating learning, self reasoning,
Those types of computations has applications
for almost every industry where humans
engage in dialogue, ask questions, test ideas
make decisions in fields such as healthcare,
finance, education, law, government services
and commerce.
25.
26. 1 mm3 of cortex:
50,000 neurons
10000 connections/neuron
(=> 500 million connections)
4 km of axons
whole brain (2 kg):
1011 neurons
1015 connections
8 million km of axons
1 mm2 of a CPU:
1 million transistors
2 connections/transistor
(=> 2 million connections)
.002 km of wire
whole CPU:
109 transistors
2*109 connections
2 km of wire
The human brain connectivity vs
CPUs
27. According to many scientists , they consider the
human brain is the most complex system in the
Universe. Exploring it is even more difficult than
exploring the space, or the deep ocean.
The main groups of functions that characterize it, are:
-1- Cognition, VS Artificial
Cognition
- 2-Intelligence VS Artificial
Intelligence
- 3-consciousness
-4-unconsciousness. VS Artificial
Consciousness
28. Many researchers believe that heart leads brain and each
heart cell has a memory!, also many researchers believe
that heart cells store information.
Information flows from heart to inside brain through special
paths, as these information leads brain cells to be able to
understand and realize , nowadays scientists are working to
establish many centers concerned about studying
relationship between heart and brain and the relation
between heart and cognitive and psychological operations.
29.
30. The six hat thinking
The human control the cognition
process, such as selecting the
proper process, the context, the
duration of processing, the need
to transit to another process and
so on.
Also, this controller need some
inputs from the environment
surrounded, the previous
experience, and the objectives
or the intentions.
One may say that controller
31. • Human Intelligence is the Cognitive ability → the
ability to perform well in cognitive tasks.
• The ability to use knowledge, solve problems,
understand complex ideas, learn quickly, and adapt
to environmental challenges.
• Human brain needs cognation to perform
intelligent tasks, while computer perform it
without cognition, that is why the computer have
limited intelligent tasks that can be done with
limited accuracy.
32. We can try to defined cognition as :
The processing, and fusion, of
knowledge from multiresources,
using an adaptive controller with
many inputs/outputs.
33. The most important part in human brain that
accomplish most work solve many problems,
generate rules form the input knowledge of
different sources.
You can easily discover the rule of tacit
knowledge, if you choose, do, deicide, say,
some thing, and you don't have a specific
explanation of that thing.
This hidden knowledge we need to know more
34.
35.
36. cognitive systems—are a category of
technologies that uses natural language
processing and machine learning to
enable people and machines to interact
more naturally to extend and magnify
human expertise and cognition.
Also, Cognitive Systems can be defined as
“Systems whose behaviour changes in an
adaptive and proactive manner in response to and
in anticipation of changes in the environment,
user characteristics and goals”.
37. • It was published at J.G. Taylor, “Cognitive computation,” Cogn.
Comput, vol.1, pp.4–16 (2009).
• Taylor raised a number of very interesting points in his attempts to
construct an artificial being empowered with its own cognitive
powers:
• Taylor’s proposal is one of very few attempts to construct a
global brain theory of cognition and consciousness.
• It is based on a unique multi-modal approach that takes into
consideration vision and attention, motor action, language and
emotion.
38. • Taylor asked a number of questions
• What is human cognition in general, and how can it be
modelled?
• What are the powers of animal cognition, and how can they
be modelled?
• How important is language in achieving a cognitive
machine, and how might it be developed in such a machine?
• What are the benchmark problems that should be able to be
solved by a cognitive machine?
39. • Does a cognitive machine have to be built in
hardware?
• How can hybridisation help in developing truly
cognitive machines?
• Is consciousness crucial?
• How are the internal mental states of others to
be discerned?
40. Scientists have created by far
the most advanced
neuromorphic (brain-like)
computer chip to date.
The chip, called TrueNorth,
consists of 1 million
programmable neurons and 256
million programmable synapses
across 4096 individual
neurosynaptic cores.
Built on Samsung’s 28nm
process and with a monstrous
16 of IBM’s finest
TrueNorth chips
(probably one of
the most expensive
motherboards in the
world)
40
41. SpiNNaker Machine
SpiNNaker is a massively-parallel neuromorphic
computing architecture designed to model large,
biologically plausible, spiking neural networks.
BrainScaleS project
Neuromorphic hardware is based on wafer-scale analog
VLSI. Each wafer implements ~200,000 spiking neurons
and 49 million synapses.
Neurogrid
Brains in Silicon group at Stanford University has built a
board with 16 neuromorphic processors that implements
1 million spiking neurons.
42. Blue Brain Project – International
researchers using an IBM 'Blue Gene'
supercomputer (thus the name Blue Brain),
are reconstructing brains of different
species; including the human brain, in
silicon.
Chief scientist Henry Markram predicts that
with Moore's Law fast-forwarding computer
technologies, a full-scale human brain
simulation of 86 billion neurons will be
43.
44. IBM hosted the Cognitive Systems Colloquium at the
T.J. Watson Research Center in Yorktown Heights,
N.Y., on Oct. 2, 2013.
This cognitive system is yielding capabilities such as
recall, learning, judgement, reasoning and inference.
Their focus is on expanding these capabilities to
recognise emotions, be more expressive in
generating speech, add perception and creativity, as
well as expanding beyond English text to multiple
languages, images and other senses.
45. 90 x IBM Power 750 servers
2880 POWER7 cores
POWER7 3.55 GHz chip
500 GB per sec on-chip bandwidth
10 Gb Ethernet network
15 Terabytes of memory
20 Terabytes of disk, clustered
Can operate at 80 Teraflops
Runs IBM DeepQA software
Scales out with and searches vast amounts of
unstructured information with UIMA & Hadoop
open source components
Linux provides a scalable, open platform,
optimized
to exploit POWER7 performance
10 racks include servers, networking, shared
46. Initial
Question
Hypothesis
Generation
Hypothesis
& Evidence
Scoring
Final Confidence
Merging & Ranking
Synthesis
Question
& Topic
Analysis
Hypothesis
Generation
Hypothesis and
Evidence Scoring
Learned Models
help combine and
weigh the Evidence
Evidence Sources
Answer
Scoring
Deep
Evidence
Scoring
Evidence
Retrieval
Answer Sources
Primary
Search
Candidate
Answer
Generation
Question
Decomposition
Hypothesis
Generation
Hypothesis and Evidence
Scoring
model
model
model
model
model
model
model
model
model
Answer &
Confidence
49. Hypothesis
& Evidence
Scoring
Final Confidence
Merging & Ranking
Synthesis
Hypothesis
Generation
Hypothesis
Generation
Learned Models
help combine and
weigh the Evidence
model
model
model
model
model
model
model
model
model
Answer &
Confidence
Hypothesis
Generation
Question
& Topic
Analysis
Answer Sources
Primary
Search
Candidate
Answer
Generation
Question
Decomposition
Initial
Question
Using models
on the merged
hypotheses,
Watson can
weigh evidence
based on prior
“experiences”
9
Once Watson has
ranked its answers, it
then provides its
answers as well as the
confidence it has in
each answer.
10
50. Jeopardy Challenge
In January 2011 Watson competed against two of
the best Jeopardy Champions (American tv
game show), it makes a good score and bit them.
There is now a new a bout preparing Waton to
pass the first year exam of medical college.
Watosn APIs, are now available for the
researcher, and testers.
53. Spiking Neuron Networks (SNNs) are often
referred to as the 3rd generation of neural
networks. Highly inspired from natural computing
in the brain and recent advances in
neurosciences,
Spiking neural networks (SNN) exhibit interesting
properties that make them particularly suitable
for applications that require fast and efficient
computation and where the timing of input/output
signals carries important information.
54. Invariant to geometrical transformations
Fixed structure of neural network
Learning – free
PCNN Properties
One-layer, two dimension NN
Lateral connection of weights
The PCNN structure is the same
as the structure of the input
object matrix S
55. Structure of PCNN neuron
Primary and Linking input
Linking part
Pulse generator
56. ijF
F
ijijij nYWVenFSnF ))1(()1()( 1
ijL
L
ijij nYWVenLnL ))1(*()1()( 2
Feeding input:
Linking input
Input part
Internal activity of neuron: ))(1()()( nLnFnU ijijij
Linking part
Output:
Threshold
potential:
{
)(if1
otherwise0
)(
nnU
nY ijij
ij
)1()1()(
nYVenn ijijij
Pulse generator
image pixel intensity iteration step W1, W2: weight matrix
VL, VF,, Vq :coefficients of potentialsL , F : decay coefficients
linking coefficientactivated neuron
non-activated
neuron
Mathematical model of PCNN neuron
57. Features generation by PCNN
0
2000
4000
6000
8000
10000
12000
0 5 10 15 20 25 30 35 40 45 50
iterations, n
G(n)
PCNN output
ij
ij nYnG )()(
input
image
PCNN output in
3. iteration step
vector of generated
features
generated feature
in 28. iteration step
58. PCNN can do erosion, dilation, thinning,
denoising, segmentation, and more
59. Cognitive Augmentation
AugCog aime to design
closed-loop systems to
modulate information flow
with respect to the user's
cognitive capacity.
Australian Art-Performer
Stelarc has a third arm
which he can control using
his abdominal muscles
60. Cognitive Radios
• A Cognitive radio is an intelligent wireless communication
system that: enhance the control process by adding
• Intelligent, autonomous control of the radio
• An ability to sense the environment
• Goal driven operation
• Processes for learning about
environmental parameters
• Awareness of its environment
• Signals
• Channels
• Awareness of capabilities of the radio
• An ability to negotiate waveforms with other radios
61. Cognitive Networking
• Cognitive networks is Scalable
autoconfiguration & network
management
• Dynamic network layer supporting
tailored functionality (IP, group
messaging, rich queries, etc.)
• Builds on the foundation of cognitive
radios, but extends it further up the
protocol stack, and explores across
stack
62.
63. HIT is a computer/phone interface that can
interact in a natural way with the user,
accept natural input in form of:
• speech and sound commands; text
commands;
• visual input, reading text (OCR),
recognizing gestures, lip movement;
HIT should have a robust understanding of
user
intentions for selected applications.
HIT should respond and behave in a
natural way.
67. What is next?
Consciousness
• “Consciousness poses the most baffling
problems in the science of the mind. There is
nothing that we know more intimately than
conscious experience, but there is nothing
that is harder to explain.” -Chalmers
• Consciousness is the quality or state of
awareness, or, of being aware of an external
object or something within oneself
68. Intelligence, consciousness and
cognitive
Human intelligence is the intellectual capacity
of humans, which is characterized by perception,
consciousness, self-awareness, and volition.
Through their intelligence, humans possess the
cognitive abilities to learn, form concepts,
understand, apply logic, and reason, including
the capacities to recognize patterns,
comprehend ideas, plan, problem solve, make
decisions, retaining, and use language to
communicate. Intelligence enables humans to
experience and think
69.
70. Artificial Consciousness
The functions of consciousness suggested by Bernard Baars :
Definition and Context Setting
Adaptation and Learning
Anticipation Function
Prioritizing and Access-Control
Decision-making or Executive Function
Analogy-forming Function
Metacognitive and Self-monitoring Function
Autoprogramming and Self-maintenance Function
Definitional and Context-setting Function.
71. What we are aware of…
The complexities
of cognition are
usually hidden
from our
consciousness.