Cognitive tics Project Report

8/3/2019 Cognitive tics Project Report

1/24

VISVESVARAYA TECHNOLOGICAL UNIVERSITY

Jnana Sangama, Belgaum590 018

ASeminar Report

On

Cognitive Informatics

Submitted by

NEHA KAPOOR

USN:5ZC10SNZ16

M.Tech in Computer Network Engineering

Under the Guidance of

Mr. Rampur Srinath

Associate Professor

DEPARTMENT OF PG STUDIES IN

COMPUTER ENGINEERING & APPLICATIONS

National Institute of Engineering, Mysore

November 2011


2/24

VISVESVARAYA TECHNOLOGICAL UNIVERSITY

Jnana Sangama, Belgaum590 018

Department of PG Studies in

Computer Engineering & Applications

NIE, Mysore570008

CERTIFICATE

Certified that the seminar report entitled Cognitive Informatics is carried out and presented by NEHA KAPOOR

with USN 5ZC10SNZ16 in partial fulfillment of requirements for the award of M.Tech (Computer Network

Engineering) by The National Institute of Engineering, Mysore during the academic year 2011 - 2012. It is certified

that all corrections and suggestions indicated by the guide have been incorporated in the seminar report.

Seminar Guide

Mr. Rampur SrinathAssociate Professor

Department of PG Studies in Computer Engineering & Applications

The National Institute of Engineering

Mysore570008

Signature of HOD Signature of Guide


3/24

Abstract

This keynote lecture presents a set of the latest advances in cognitive informatics (CI) that leads

to the design and implementation of future generation computers known as the cognitivecomputers that are capable of thinking and feeling. The theory and philosophy behind the nextgeneration computers and computing technologies are CI. The theoretical framework of CI maybe classified as an entire set of cognitive functions and processes of the brain and an enriched setof descriptive mathematics, the cognitive computers are created for cognitive and perceptibleconcept/knowledge processing based on contemporary mathematics such as concept algebra,real-time process algebra, and system algebra. Because the cognitive computers implement thefundamental cognitive processes of the natural intelligence such as the learning, thinking, formalinference, and perception processes, they are novel information processing systems that thinkand feel. The cognitive computers are centered by the parallel inference engine and perceptionengine that implement autonomic learning/reasoning and perception mechanisms based on

descriptive mathematics.


4/24

Table of ContentsIntroduction ................................................................................................................................................................... 5Cognitive Informatics .................................................................................................................................................... 6

The Science of Abstract Intelligence and Computational Intelligence .......... ........... .......... ........... .......... ........... ....... 6Theoretical framework of CI ......................................................................................................................................... 7The Fundamental Theories of CI ................................................................................................................................... 8

The Information-Matter-Energy Model ..................................................................................................................... 8The Layered Reference Model of the Brain ................. ........... .......... .......... ........... .......... ........... .......... ........... ......... 9The OAR Model of Information Representation in the Brain ................ ........... .......... ........... .......... .......... ........... .. 10The Cognitive Informatics Model of the Brain .......... .......... ........... .......... .......... ........... .......... ........... .......... ........... 10Natural Intelligence (NI) .......................................................................................................................................... 10Neural Informatics (NeI) ......................................................................................................................................... 11Cognitive Informatics Laws of Software ................................................................................................................. 11Mechanisms of Human Perception Processes .......................................................................................................... 12The Cognitive Processes of Formal Inferences .......... ........... .......... ........... .......... ........... .......... ........... .......... ......... 12The Formal Knowledge System .............................................................................................................................. 13

Denotational Mathematics for CI ................................................................................................................................ 13Applications of CI ....................................................................................................................................................... 14Cognitive Computing and Cognitive Computers ......................................................................................................... 14

Cognitive computing: thought for the future ........................................................................................................... 15The Architecture of Future Generation Computers ..................................................................................................... 16

The Problem with Modern Computers .................................................................................................................... 17Autonomic Computing ................................................................................................................................................ 18Cognitive Properties of Knowledge ............................................................................................................................. 19IBM claims cognitive computing breakthrough .......................................................................................................... 19

IBM wants to emulate that architecture with its new chips. ........... .......... ........... .......... .......... ........... .......... ........... 20Conclusions ................................................................................................................................................................. 22References ................................................................................................................................................................... 23


5/24

Introduction

The development of classical and contemporary informatics, the cross fertilization betweencomputer science, systems science cybernetics, computer/software engineering cognitivescience, knowledge engineering, an neuropsychology, has led to an entire rang of an extremelyinteresting and new research field known as Cognitive Informatics. Informatics is the science ofinformation that studies the nature of information; its processing, a ways of transformationbetween information matter, and energy.

Definition 1. Cognitive Informatics (CI) is a transdisciplinary enquiry of cognitive andInformation sciences that investigates the internal information processing mechanism andprocesses of the brain and natural intelligence, and their engineering application via aninterdisciplinary approach. In many disciplines of human knowledge almost all of the hard

problems yet to be solved share a common root in the understanding of the mechanisms ofnatural intelligence and the cognitive processes of the brain Therefore, CI is a discipline thatforges link between a number of natural science and life science disciplines with informatics andcomputing science. A number of fundamental human wonders from the context of CI, such as:How consciousness is generated as a highly complex cognitive state in human mind on the basisof physiological metabolism? How natural intelligence is generated on the basis of basicbiological a physiological structures? How intelligence functions logically and physiologically?And how natural a machine intelligence is converged on the basis of CI and computationalintelligence?

Figure 1. Cognitive Informatics


6/24

Cognitive Informatics

The Science of Abstract Intelligence and Computational Intelligence

Information is the third essence of the word supplementing energy and matter. A key discoveryin information science is the basic unit of information, bit, abbreviated from a binary digit,which forms a shared foundation of computer science and informatics. The science ofinformation, informatics, has gone through three generations of evolution, known as the classic,modern, and cognitive informatics, since Shannon proposed the classic notion of information .The classical information theory founded by Shannon (1948) defined information as aprobabilistic measure of the variability of message that can be obtained from a message source.Along with the development in computer science and in the IT industry, the domain of

informatics has been dramatically extended in the last few decades. This led to the moderninformatics that treats information as entities of messages rather than a probabilisticmeasurement of the variability of messages as in that of the classic information theory. The newperception of information is found better to explain the theories in computer science andpractices in the IT industry. However, both classic and modern views on information are onlyfocused on external information. The real sources and destinations of information, the humanbrains, are often overlooked. This leads to the third generation of informatics, cognitiveinformatics, which focuses on the nature of information in the brain, such as informationacquisition, memory, categorization, retrieve, generation, representation, and communication.Information in cognitive informatics is defined as the abstract artifacts and their relations that canbe modeled, processed, stored and processed by human brains. Cognitive informatics is emerged

and developed based on the multidisciplinary research in cognitive science, computing science,information science, abstract intelligence, and denotational mathematics.CI is a cutting-edge and multidisciplinary research area that tackles the fundamental problemsshared by modern informatics, computation, software engineering, AI, cybernetics, cognitivescience, neuropsychology, medical science, philosophy, linguistics, brain sciences, and manyothers. The development and the cross fertilization among the aforementioned science andengineering disciplines have led to a whole range of extremely interesting new research areas.


7/24

Figure 2. Cognitive Informatics

Theoretical framework of CI

The structure of the theoretical framework of CI is described in Figure 3, which covers theInformation-Matter-Energy (IME) model, The Layered Reference Model of the Brain (LRMB),The Object-Attribute-Relation (OAR) model of information representation in the brain, Thecognitive informatics model of the brain , Natural Intelligence (NI) , Autonomic Computing(AC) , Neural Informatics (NeI) , CI laws of software, The mechanisms of human perceptionprocesses , The cognitive processes of formal inferences , and The formal knowledge system.The theoretical framework of CI is explained in the fundamental theories of CI section. Threestructures of new descriptive mathematics such as Concept Algebra (CA), Real-Time ProcessAlgebra (RTPA), and System Algebra (SA) are introduced in the denotational mathematics forCI in order to rigorously deal with knowledge and cognitive information representation and

manipulation in a formal and coherent framework.


8/24

Figure 3. Theoretical framework of CI

The Fundamental Theories of CI

The fundamental theories of CI encompass 10 transdisciplinary areas and fundamental models.This section presents an intensive review of the theories developed in CI, which form afoundation for exploring the natural intelligence and their applications in brain science, neuralinformatics, computing, knowledge engineering, and software engineering.

The Information-Matter-Energy Model

Figure 4. I-M-E model

Information is recognized as the third essence of the natural world supplementing to matter andenergy, because the primary function of the human brain is information processing.

Theorem 1. A generic worldview, the IME model states that the natural world (NW) that formsthe context of human beings is a dual world: one aspect of it is the physical or the concrete world(PW), and the other is the abstract or the perceptive world (AW), where matter (M) and energy(E) are used to model the former, and information (I) to the latter, that is:NW PW || AW= p (M, E)|| a (I) (1)= n (I, M, E)

where || denotes a parallel relation, and p, a, and n are functions that determine a certain PW,AW,orNW, respectively.According to the IME model, information plays a vital role in connecting the physical world withthe abstract world. Models of the natural world have been well studied in physics and othernatural sciences. However, the modeling of the abstract world is still a fundamental issue yet tobe explored in cognitive informatics, computing, software science, cognitive science, brainsciences, and knowledge engineering. Especially the relationships between I-M-E and theirtransformations are deemed as one of the fundamental questions in CI.


9/24

Corollary 1. The natural worldNW(I, M, E), particularly part of the abstract world,AW(I), iscognized and perceived differently by individuals because of the uniqueness of perceptions andmental contexts among people.Corollary 1 indicates that although the physical world PW(M, E) is the same to everybody,

the natural worldNW(I, M, E) is unique to different individuals because the abstract worldAW(I), as a part of it, is subjective depending on the information an individual obtains andperceives.

The Layered Reference Model of the Brain

Figure 5. LRMB model

The LRMB is developed to explain the fundamental cognitive mechanisms and processes ofnatural intelligence. Because a variety of life functions and cognitive processes have beenidentified in CI, psychology, cognitive science, brain science, and neurophilosophy, there is aneed to organize all the recurrent cognitive processes in an integrated and coherent framework.The LRMB model explains the functional mechanisms and cognitive processes of naturalintelligence that encompasses 37 cognitive processes at six layers known as the sensation,memory, perception, action, metacognitive, and higher cognitive layers from the bottom-up.


10/24

LRMB elicits the core and highly repetitive recurrent cognitive processes from a huge variety oflife functions, which may shed light on the study of the fundamental mechanisms andinteractions of complicated mental processes, particularly the relationships and interactionsbetween the inherited and the acquired life functions as well as those of the subconscious andconscious cognitive processes.

The OAR Model of Information Representation in the Brain

Investigation into the cognitive models of information and knowledge representation in the brainis perceived to be one of the fundamental research areas that help to unveil the mechanisms ofthe brain. The Object- Attribute-Relation (OAR) model (describes human memory, particularlythe long-term memory, by using the relational metaphor, rather than the traditional containermetaphor that used to be adopted in psychology, computing, and information science. The OARmodel shows that human memory and knowledge are represented by relations, that is,connections of synapses between neurons, rather than by the neurons themselves as the

traditional container metaphor described. The OAR model can be used to explain a wide range ofhuman information processing mechanisms and cognitive processes.

The Cognitive Informatics Model of the Brain

The human brain and its information processing mechanisms are centered in CI. A cognitiveinformatics model of the brain is proposed, which explains the natural intelligence viainteractions between the inherent (subconscious) and acquired (conscious) life functions. Themodel demonstrates that memory is the foundation for any natural intelligence. Formalism informs of mathematics, logic, and rigorous treatment is introduced into the study of cognitive and

neural psychology and natural informatics. Fundamental cognitive mechanisms of the brain, suchas the architecture of the thinking engine, internal knowledge representation, long-term memoryestablishment, and roles of sleep in long-term memory development have been investigated.

Natural Intelligence (NI)

Natural Intelligence (NI) is the domain of CI. Software and computer systems are recognized asa subset of intelligent behaviors of human beings described by programmed instructiveinformation. The relationship between Artificial Intelligence (AI) and NI can be describedby the following theorem.

Theorem 3. The law of compatible intelligent capability states that artificial intelligence (AI) isalways a subset of the natural intelligence (NI), that is:AI NI (5)Theorem 3 indicates that AI is dominated by NI. Therefore, one should not expect a computer ora software system to solve a problem where humans cannot. In other words, no AI or computingsystem may be designed and/or implemented for a given problem where there is no solutionbeing known by human beings.


11/24

Neural Informatics (NeI)

Definition 5. Neural Informatics (NeI) is a new interdisciplinary enquiry of the biological andphysiological representation of information and knowledge in the brain at the neuron level andtheir abstract mathematical models. NeI is a branch of CI, where memory is recognized as thefoundation and platform of any natural or artificial intelligence.

Definition 6. The Cognitive Models of Memory (CMM) states that the architecture of humanmemory is parallel configured by the Sensory Buffer Memory (SBM), Short-Term Memory(STM), Long-Term Memory (LTM), and Action- Buffer Memory (ABM), that is:CMM SBM|| STM|| LTM

|| ABMwhere the ABM is newly identified. The major organ that accommodates memories in the brainis the cerebrum or the cerebral cortex. In particular, the association and premotor cortex in thefrontal lobe, the temporal lobe, sensory cortex in the frontal lobe, visual cortex in the occipitallobe, primary motor cortex in the frontal lobe, supplementary motor area in the frontal lobe, andprocedural memory in cerebellum. The CMM model and the mapping of the four types of humanmemory onto the physiological organs in the brain reveal a set of fundamental mechanisms ofNeI. The OAR model of information/knowledge representation described in the OAR model ofinformation representation in the brain section provides a generic description ofinformation/knowledge representation in the brain.The theories of CI and NeI explain a number of important questions in the study of NI.

Enlightening conclusions derived in CI and NeI are such as: (a) LTM establishment is asubconscious process; (b) The long-term memory is established during sleeping; (c) The majormechanism for LTM establishment is by sleeping; (d) The general acquisition cycle of LTM isequal to or longer than 24 hours; (e) The mechanism of LTM establishment is to update theentire memory of information represented as an OAR model in the brain; and (f) Eye movementand dreams play an important role in LTM creation. The latest development in CI and NeI haveled to the determination of the magnificent and expected capacity of human memory as describedin the Estimation of the Capacity of Human Memory section.

Cognitive Informatics Laws of Software

It is commonly conceived that software as an artifact of human creativity is not constrained bythe laws and principles discovered in the physical world. However, it is unknown whatconstrains software. The new informatics metaphor proposed by the author in CI perceivessoftware is a type of instructive and behavioral information. Based on this, it is asserted thatsoftware obeys the laws of informatics. A comprehensive set of 19 CI laws for software havebeen established such as:


12/24

1. Abstraction2. Generality3. Cumulativeness4. Dependency on cognition5. Three-dimensional behavior space known as the object (O), space (S), and time (T)

6. Sharability7. Dimensionless8. Weightless9. Transformability between I-M-E10.Multiple representation forms11. Multiple carrying media12. Multiple transmission forms13. Dependency on media14. Dependency on energy15. Wearless and time dependency16. Conservation of entropy

17. Quality attributes of informatics18. Susceptible to distortion19. ScarcityThe informatics laws of software extend the knowledge on the fundamental laws and propertiesof software where the conventional product metaphor could not explain. Therefore, CI forms oneof the foundations of software engineering and computing science.

Mechanisms of Human Perception Processes

Definition 7. Perception is a set of interpretive cognitive processes of the brain at thesubconscious cognitive function layers that detects, relates, interprets, and searches internal

cognitive information in the mind. Perception may be considered as the sixth sense of humanbeings, which almost all cognitive life functions rely on. Perception is also an importantcognitive function at the subconscious layers that determines personality. In other words,personality is a faculty of all subconscious life functions and experience cumulated via consciouslife functions. According to LRMB, the main cognitive processes at the perception layer areemotion, motivation, and attitude. The relationship between the internal emotion, motivation,attitude, and the embodied external behaviors can be formally and quantitatively described by themotivation/attitude-driven behavioral (MADB) model which demonstrates that complicatedpsychological and cognitive mental processes may be formally modeled and rigorously describedby mathematical means.

The Cognitive Processes of Formal Inferences

Theoretical research is predominately an inductive process, while applied research is mainly adeductive one. Both inference processes are based on the cognitive process and means ofabstraction. Abstraction is a powerful means of philosophy and mathematics. It is also apreeminent trait of the human brain identified in CI studies. All formal logical inferences and


13/24

reasoning can only be carried out on the basis of abstract properties shared by a given set ofobjects under study.

Definition 8. Abstraction is a process to elicit a subset of objects that shares a common propertyfrom a given set of objects and to use the property to identify and distinguish the subset from the

whole in order to facilitate reasoning. Abstraction is a gifted capability of human beings.Abstraction is a basic cognitive process of the brain at the metacognitive layer according toLRMB. Only by abstraction can important theorems and laws about the objects under study beelicited and discovered from a great variety of phenomena and empirical observations in an areaof inquiry.

Definition 9. Inferences are a formal cognitive process that reasons a possible causality fromgiven premises based on known causal relations between a pair of cause and effect proven trueby empirical arguments, theoretical inferences, or statistical regulations. Formal inferences maybe classified into the deductive, inductive, abductive, and analogical categories. Deduction is acognitive process by which a specific conclusion necessarily follows from a set of general

premises. Induction is a cognitive process by which a general conclusion is drawn from a set ofspecific premises based on three designated samples in reasoning or experimental evidences.Abduction is a cognitive process by which an inference to the best explanation or most likelyreason of an observation or event. Analogy is a cognitive process by which an inference aboutthe similarity of the same relations holds between different domains or systems, and/or examinesthat if two things agree in certain respects, and then they probably agree in others. A summary ofthe formal definitions of the five inference techniques.

The Formal Knowledge System

Mathematical thoughts provide a successful paradigm to organize and validate humanknowledge, where once a truth or a theorem is established, it is true until the axioms orconditions that it stands for are changed or extended. A proven truth or theorem in mathematicsdoes not need to be argued each time one uses it. This is the advantage and efficiency of formalknowledge in science and engineering. In other words, if any theory or conclusion may beargued from time-to-time based on a wiser idea or a trade-off, it is an empirical result rather thana formal one.

Denotational Mathematics for CI

The history of sciences and engineering shows that new problems require new forms ofmathematics. CI is a new discipline, and the problems in it require new mathematical means thatare descriptive and precise in expressing and denoting human and system actions and behaviors.Conventional analytic mathematics are unable to solve the fundamental problems inherited in CIand related disciplines such as neuroscience, psychology, philosophy, computing, softwareengineering, and knowledge engineering. Therefore, denotational mathematical structures andmeans beyond mathematical logic are yet to be sought. Three types of new mathematics,


14/24

Concept Algebra (CA), Real-Time Process Algebra (RTPA), and System Algebra (SA), arecreated in CI to enable rigorous treatment of knowledge representation and manipulation in aformal and coherent framework. The three new structures of contemporary mathematics haveextended the abstract objects under study in mathematics from basic mathematical entities ofnumbers and sets to a higher level, that is, concepts, behavioral processes, and systems. A wide

range of applications of the denotational mathematics in the context of CI has been identified.

Definition 11. Concept algebra is a new mathematical structure for the formal treatment ofabstract concepts and their algebraic relations, operations, and associative rules for composingcomplex concepts and knowledge.

Definition 14. Real-Time Process Algebra is a set of formal notations and rules for describingalgebraic and real-time relations of software processes.Definition 19. System algebra is a new abstract mathematical structure that provides an algebraictreatment of abstract systems as well as their relations and operational rules for forming complexsystems.

Applications of CI

The last two sections have reviewed the latest development of fundamental researches in CI,particularly its theoretical framework and descriptive mathematics. A wide range of applicationsof CI has been identified in multidisciplinary and transdisciplinary areas, such as: (1) Thearchitecture of future generation computers; (2) Estimation the capacity of human memory; (3)Autonomic computing; (4) Cognitive properties of information, data, knowledge, and skills inknowledge engineering; (5) Simulation of human cognitive behaviors using descriptive

mathematics; (6) Agent systems; (7) CI foundations of software engineering; (8) Deductivesemantics of software; and (9) Cognitive complexity of software.

Cognitive Computing and Cognitive Computers

Computing systems and technologies can be classified into the categories of imperative,autonomic, and cognitive computing from the bottom up. The imperative computers are atraditional and passive system based on stored-program controlled behaviors for data processing.The autonomic computers are goal-driven and self-decision-driven machines that do not rely oninstructive and procedural information. Cognitive computers are more intelligent computers

beyond the imperative and autonomic computers, which embody major natural intelligencebehaviors of the brain such as thinking, inference, and learning.

Definition 10. A cognitive computer is an intelligent knowledge processor with the capabilitiesof autonomous inference and perception that mimics the mechanisms of the brain. Thearchitectural model of cognitive computers can be refined by a behavioral model that evolvescomputing technologies from the conventional imperative behaviors to the autonomic andcognitive behaviors.


15/24

Definition 11. The behavioral model of a cognitive computer is an abstract logical model ofcomputing platform denoted by a set of parallel computing architectures and behaviors.Cognitive computers are aimed at cognitive and perceptive concept/knowledge processing basedon contemporary denotational mathematics. As that of mathematical logic and Boolean algebra

are the mathematical foundations of von Neumann architectures, the mathematical foundationsof cognitive computers are based on denotational mathematics. According to the LRMBreference model, since all the 39 fundamental cognitive processes of human brains can beformally described in concept algebra and RTPA, they are simulatable and executable by thecognitive computers.

Cognitive computing: thought for the future

Figure 6. Cognitive Computing

A cognitive computing system monitoring the world's water supply could contain a network ofsensors and actuators that constantly record and report metrics such as temperature, pressure,wave height, acoustics and ocean tide, and issue tsunami warnings based on its decision making.Researchers at IBM have been working on a cognitive computing project called Systems ofNeuromorphic Adaptive Plastic Scalable Electronics (SyNAPSE). By reproducing the structure

and architecture of the brainthe way its elements receive sensory input, connect to each other,adapt these connections, and transmit motor outputthe SyNAPSE project models computingsystems that emulate the brain's computing efficiency, size and power usage without beingprogrammed.IBM is combining principles from nanoscience, neuroscience and supercomputingas part of a multi-year cognitive computing initiative. The Defense Advanced Research ProjectsAgency (DARPA) has awarded approximately US$21 million in new funding for phase 2 of theSyNAPSE project. For this project, a world-class, multi-dimensional team has been assembled,


16/24

consisting of IBM researchers and collaborators from Columbia University; Cornell University;University of California, Merced; and University of Wisconsin-Madison.

The Architecture of Future Generation Computers

Conventional machines are invented to extend human physical capability, while moderninformation processing machines, such as computers, communication networks, and robots, aredeveloped for extending human intelligence, memory, and the capacity for informationprocessing. Recent advances in CI provide formal description of an entire set of cognitiveprocesses of the brain. The fundamental research in CI also creates an enriched set ofcontemporary denotational mathematics, for dealing with the extremely complicated objects andproblems in natural intelligence, neural informatics, and knowledge manipulation. The theoryand philosophy behind the next generation computers and computing methodologies are CI. It is

commonly believed that the future-generation computers, known as the cognitive computers, willadopt non-von Neumann (von Neumann, 1946) architectures. The key requirements forimplementing a conventional stored-program controlled computer are the generalization ofcommon computing architectures and the computer is able to interpret the data loaded in memoryas computing instructions. These are the essences of stored-program controlled computers knownas the von Neumann (1946) architecture. Von Neumann elicited five fundamental and essentialcomponents to implement general-purpose programmable digital computers in order to embodythe concept of stored-program-controlled computers.

Definition 22. A von Neumann Architecture (VNA) of computers is a 5-tuple that consistsof the components: (a) the arithmetic-logic unit (ALU), (b) the control unit (CU) with a program

counter (PC), (c) a memory (M), (d) a set of input/output (I/O) devices, and (e) a bus (B) thatprovides the data path between these components, that is:VNA (ALU, CU, M, I/O, B)

Definition 23. Conventional computers with VNA are aimed at stored-program-controlled dataprocessing based on mathematical logic and Boolean algebra. A VNA computer is centric by thebus and characterized by the all purpose memory for both data and instructions. A VNA machineis an extended Turing machine (TM), where the power and functionality of all components ofTM including the control unit (with wired instructions), the tape (memory), and the head of I/O,are greatly enhanced and extended with more powerful instructions and I/O capacity.


17/24

Figure 7.Architecture of cognitive machine

Definition 24. A Wang Architecture (WA) of computers, known as the Cognitive Machine, is aparallel structure encompassing an Inference Engine (IE) and a Perception Engine (PE), that is:WA (IE || PE)

= (KMU// The knowledge manipulation unit || BMU//The behavior manipulation unit || EMU //The experience manipulation unit || SMU// The skill manipulation unit) || (BPU // The behaviorperception unit EPU // The experience perception unit)WA computers are not centered by a CPU for data manipulation as the VNA computers do. TheWA computers are centered by the concurrent IE and PE for cognitive learning and autonomicperception based on abstract concept inferences and empirical stimuli perception. The IE isdesigned for concept/knowledge manipulation according to concept algebra particularly the nineconcept operations for knowledge acquisition, creation, and manipulation. The PE is designedfor feeling and perception processing according to RTPA and the formally described cognitiveprocess models of the perception layers as defined in the LRMB model.

The Problem with Modern Computers

For the past half-century, most computers run on whats known as von Neumann architecture,and cognitive computers run on non-von Neumann architecture. In a von Neumann system, theprocessing of information and the storage of information are kept separate. Data travels to andfrom the processor and memorybut the computer cant process and store at the same time. By


18/24

the nature of the architecture, its a linear process. Thats why software is written as a set ofinstructions for a computer to followits a linear sequence of events, built for a linear process.This is where clock speed comes in the faster the clock speed the faster the computer canprocess those linear instructions.According to Dr. Modha, von Neumann architecture was essential to develop computers in the

days of vacuum tubes and early transistors, but modern chip building techniques have exposedits limitations. Weve gotten to the point where we can pack more transistors on a chip than wecan actually power, because if we powered them all, theyd burn out due to the excess heat

created by the electricity in the chip.The genius of the von Neumann architecture back then is the weakness of today's computers.There four basic weaknesses in the von Neumann architecture that inhibits modern computersfrom achieving greater speed and power. The first one is the fact that instructions and data aredistinguished only implicitly through usage. As he points out, the higher level languagescurrently used for programming make a clear distinction between the instructions and the dataand have no provision for executing data or using instructions as data.The second is that the memory is a single memory, sequentially addressed.

The third weakness is that the memory is one-dimensional. These are in conflict with ourprogramming languages. Most of the resulting program, therefore, is generated to provide for themapping of multidimensional data onto the one dimensioned memory and to contend with theplacement of all of the data into the same memory.As for the fourth weakness, the problem is that the meaning of the data is not stored with it. Inother words, it is not possible to tell by looking at a set of bits whether that set of bits representsan integer, a floating point number or a character string. In a higher level language, we associatesuch a meaning with the data, and expect a generic operation to take on a meaning determined bythe meaning of its operands. The implications of these weaknesses are significant.One facet of this is the fundamental view of memory as a "word at a time" kind of device. Aword is transferred from memory to the CPU or from the CPU to memory. All of the data, thenames (locations) of the data, the operations to be performed on the data, must travel betweenmemory and CPU a word at a time. This bottleneck is not only a physical limitation, but hasserved also as an "intellectual bottleneck" in limiting the way we think about computation andhow to program it.

Autonomic Computing

The approaches to implement intelligent systems can be classified into those of biologicalorganisms, silicon automata, and computing systems. Based on CI studies, autonomic computing

is proposed as a new and advanced computing technique built upon the routine, algorithmic, andadaptive systems. The approaches to computing can be classified into two categories known asimperative and autonomic computing. Corresponding to these, computing systems may beimplemented as imperative or autonomic computing systems.

Definition 26. An imperative computing system is a passive system that implementsdeterministic, context-free, and stored-program controlled behaviors.


19/24

Definition 27. An autonomic computing system is an intelligent system that autonomouslycarries out robotic and interactive actions based on goal- and event-driven mechanisms. Theimperative computing system is a traditional passive system that implements deterministic,context-free, and stored-program controlled behaviors, where a behavior is defined as a set ofobservable actions of a given computing system. The autonomic computing system is an active

system that implements nondeterministic, context-dependent, and adaptive behaviors, which donot rely on instructive and procedural information, but are dependent on internal status andwillingness that formed by long-term historical events and current rational or emotional goals.

Cognitive Properties of Knowledge

Almost all modern disciplines of science and engineering deal with information and knowledge.According to CI theories, cognitive information may be classified into four categories known asknowledge, behaviors, experience, and skills.

Definition 28. The taxonomy of cognitive information is determined by its types of inputs andoutputs to and from the brain during learning and information processing, where both inputs andoutputs can be either abstract information (concept) or empirical information (actions). It isnoteworthy that the approaches to acquire knowledge/behaviors and experience/ skills arefundamentally different. The former may be obtained either directly based on hands-on activitiesor indirectly by reading, while the latter can never be acquired indirectly. The followingimportant conclusions on information manipulation and learning for both human and machinesystems can be derived.

Theorem 9. The principle of information acquisition states that there are four sufficientcategories of learning known as those of knowledge, behaviors, experience, and skills.

Theorem 9 indicates that learning theories and their implementation in autonomic and intelligentsystems should study all four categories of cognitive information acquisitions, particularlybehaviors, experience, and skills rather than only focusing on knowledge.

Corollary 3. All the four categories of information can be acquired directly by an individual.

Corollary 4. Knowledge and behaviors can be learnt indirectly by inputting abstract information,while experience and skills must be learned directly by hands-on or empirical actions.

IBM claims cognitive computing breakthrough

The so-called cognitive chips have two prototypes that are currently being tested. Thesemiconductors were created out of standard technology in IBMs fabrication plant. Both cores

were fabricated in a 45 nanometer process and feature 256 neurons. One core contains 262,144programmable synapsesbasically the social network in the chipand 65,536 learningsynapses. IBM has demonstrated navigation, machine vision, pattern recognition, associativememory and classification with these chips. What do these cognitive chips add up to?


20/24

Dharmendra Modha, a project leader for IBM Research, said these new prototype chips can leadto systems that complement todays computers. Cognitive computers would be more about

processing unstructured data and various inputs. These cognitive chips can create a newgeneration of computers to complement todays. Todays computer would be left brainfast,analytical, rational and structuredthe cognitive side would be the other side, which is slow,

low power and unstructured its right brain to left brain, explained Modha. Bringing thistechnology forward completes the computing tool chest.In many respects, the brainwhichpacks a lot of low power computational heft in a tight spaceis the Holy Grail of computing.The upshot is that these chips could be cobbled together to correlate data, create hypotheses andremember things. The product of these chips would be a cognitive computer, according to IBM.IBM sees multiple applications for these cognitive computing systems, which would fit in thesize of a shoebox. Among potential uses:

Computers that could take in inputs such as texture, smell and feel to gauge whether foodwas outdated.

Financial applications to monitor trading and recognize patterns in a way todaysalgorithms cant.

Traffic monitoring. And system monitoring for waterways and other natural resources.

This new computing unit, or core, is analogous to the brain. It has neurons, or digital

processors that compute information. It has synapses which are the foundation of learning andmemory. And it has axons, or data pathways that connect the tissue of the computer.

While it sounds simple enough, the computing unit is radically different from the way mostcomputers operate today. Modern computers are based on the von Neumann architecture, namedafter computing pioneer John von Neumann and his work from the 1940s.In von Neumann

machines, memory and processor are separated and linked via a data pathway known as a bus.Over the past 65 years, von Neumann machines have gotten faster by sending more and moredata at higher speeds across the bus, as processor and memory interact. But the speed of acomputer is often limited by the capacity of that bus, leading some computer scientists to call itthe von Neumann bottleneck. The brain-like processors with integrated memory dont operatefast at all, sending data at a mere 10 hertz, or far slower than the 5 gigahertz computer processorsof today. But the human brain does an awful lot of work in parallel, sending signals out in alldirections and getting the brains neurons to work simultaneously. Because the brain has morethan 10 billion neuron and 10 trillion connections (synapses) between those neurons, thatamounts to an enormous amount of computing power.

IBM wants to emulate that architecture with its new chips.

The research team has built its first brain-like computing units, with 256 neurons, an array of 256by 256 (or a total of 65,536) synapses, and 256 axons. (A second chip had 262,144 synapses) Inother words, it has the basic building block of processor, memory, and communications. Thisunit, or core, can be built with just a few million transistors (some of todays fastest microchipscan be built with billions of transistors).Modha said that this new kind of computing will likelycomplement, rather than replace, von Neumann machines, which have become good at solving


21/24

problems involving math, serial processing, and business computations. The disadvantage is thatthose machines arent scaling up to handle big problems well any more. They are using too much

power and are harder to program. The more powerful a computer gets, the more power itconsumes, and manufacturing requires extremely precise and expensive technologies. And themore components are crammed together onto a single chip, the more they leak power, even in

stand-by mode. So they are not so easily turned off to save power. The advantage of the humanbrain is that it operates on very low power and it can essentially turn off parts of the brain whenthey arent in use.


22/24

Conclusions

CI has been described as a new discipline that studies the natural intelligence and internalinformation processing mechanisms of the brain, as well as processes involved in perception and

cognition. CI is a new frontier across disciplines of computing, software engineering, cognitivesciences, neuropsychology, brain sciences, and philosophy in recent years. It has beenrecognized that many fundamental issues in knowledge and software engineering are based onthe deeper understanding of the mechanisms of human information processing and cognitiveprocesses.A coherent set of theories for CI has been described in this article, such as the Information-Matter-Energy model, Layered Reference Model of the Brain, the OAR model of informationrepresentation, Natural Intelligence vs. Artificial Intelligence, Autonomic Computing vs.imperative computing, CI laws of software, mechanisms of human perception processes, thecognitive processes of formal inferences, and the formal knowledge system. Three contemporarymathematical means have been created in CI known as the denotational mathematics. Within the

new forms of denotational mathematical means for CI, Concept Algebra has been designed todeal with the new abstract mathematical structure of concepts and their representation andmanipulation in learning and knowledge engineering. Real-Time Process Algebra has beendeveloped as an expressive, easy-to-comprehend, and language-independent notation system,and a specification and refinement method for software system behaviors description andspecification. System Algebra has been created to the rigorous treatment of abstract systems andtheir algebraic relations and operations.Major problems yet to be solved in CI are such as the architectures of the brain, mechanisms ofthe natural intelligence, cognitive processes, mental phenomena, and personality. It isparticularly interested in computing and software engineering to explain the mechanisms andprocesses of memory, learning and thinking. It is expected that any breakthrough in CI will be

profoundly significant towards the development of the next generation technologies ininformatics, computing, software, and cognitive sciences.


23/24

References

[1] Yingxu Wang, Editor-in-Chief, JCiNi Cognitive Informatics: Exploring the TheoreticalFoundations for Natural Intelligence, Neural Informatics, Autonomic Computing, and Agent

Systems.

[2] A Doctrine of Cognitive Informatics (CI).

[3] Yingxu Wang University of Calgary, Canada Chapter VI The Cognitive Processes ofFormal Inferences.

[4] Yingxu Wang, University of Calgary, Canada The Theoretical Framework of CognitiveInformatics.

[5] Yingxu Wang, Senior Member, IEEE, and Ying Wang Cognitive Informatics Models of the

Brain


24/24

Documents

Cognitive tics Project Report