With all due respect to current technology computers today are just large calculators but they really don’t think . So Cognitive Computing is in simple words is how we get computers to behave,think and interact the same way as humans do .The world today always provides us with in-perfect information but cognitive systems will allow us to collect information from what is heard what is seen,what is felt and use that information to provide more accurate information .
” Watson ” the first human brain machine is poised with less programming it learns and thinks like a person . Its very good at natural language and exercise to make self a better system.
Why Cognitive Computing ??
If we look at human brain its a simple two half divided structure with cerebral cortices separated from one another.The mid way corpus acts as the communication media between the two division. Our right hemisphere acts as parallel processor and the left part as a serial processor . Each of our different hemisphere have different personality .Our Right part of the brain is concerned about the present moment , it thinks in pictures. Information in form of energy streams into our system.It makes us visualize what the present moment feels like ? smells like ? looks like ?? we as energy beams are connected as per as right hemisphere, treats us as one human family . It strives for what is real with motto ” Right here Right now” . This part of our awesome brain learns kinesthetically through movements of our body.
Whereas the left hemisphere is more into thinking and methodical approach . Its more into the past and future . It is dully designed to take the collage of details and takes details and more details of the details and separates us form the surrounding with a motto ” I am ”
Its a stone of possibility , stone of imagination where Neuroscience , super computer , nano technology meets together to create a radically different computer architecture . So that it can function ,low-power ,small size, real-time of human brain . Modern computers are thus sequential symbolic nature that can be linked to left brain . Whereas the coming era of Cognitive Computing is capable parallel short complex thinking same as the right part of our brain.
A central notion of neuroscience, sometimes called the ” neuron doctrine,” says that specialized cells in the brain, the neurons, are the biological substrate of brain computation.Human brain is sparse .Adult humans have about 100 trillion synapses infact six orders of its magnitude less than would be required to completely and directly connect the tens of billions of neurons that make up the brain.
Synapse are the media connecting the axiom of one neuron to the dendrite of the next .The transmitting media are the neurotransmitter . It is mainly of two types pre-synaptic and post synaptic.A mimic to human brain is really hard to achieve .So researchers among the globe is striving for mocking it to 4 species with brain structure simple as compared to our’s.
The cerebral fiber pathways formed a rectilinear three-dimensional grid continuous with the three principal axes of development. Cortical pathways formed parallel sheets of interwoven paths in the longitudinal and medico-lateral axes, in which major pathways were with condensations .
A brain of a worm is though achieved and the great minds in globe are looking forward for architecture same as that of macaque monkey.
LEVEL OF ABSTRACTION
Achieving what known as level of abstraction is important when comes in terms of Cognitive Computing. So What is a necessity in modular form is what solves the conundrum .
Four things mandatory to be looked upon for cognitive computing system are .
4).Ultra low power
With each core in the architecture having Neurons , Dendrites , Synapses,Axons. Connected in similar terms as that of human mind.
We don’t have any oracle to guide us through the complexity , its a real tough job to manipulate and implement what known as Cognitive Computing Systems.
How to measure Cognition
Alas a practical mono drama is not possible always , a simulation to embark and compare the architecture with that with human brain solves the problem.
So the next generation Cognitive Computing can be slowly and methodically compared with biologically inspired human brain with help of a Simulation.
This was the best that great minds across globe achieve given the available super computing resources. The Livermore National Lab with 147,456 processors and 144 TB of main memory, the simulation used 1.6 billion neurons and 9 trillion synapses.
For some scale reference points, consider this:
- A cat cortex has roughly 760 million neurons and 6 trillion synapses. So, present simulation exceeds cat-scale.
- A monkey cortex has roughly 2 billion neurons and 20 trillion synapses. So its simulation is roughly equal to 50 percent of monkey-scale.
- A human cortex has 22 billion neurons and 220 trillion synapses. So its simulation, then, is 4.5 percent of human-scale .
The Compass simulator is an all-purpose built in simulator with multi dimensional artitrectures to pursue novel architectures, algorithms, and applications. Compass is indispensable for
(a) TrueNorth correctness via regression testing
(b) TrueNorth dynamics analysis
(c) Inter-core communication topologies
(d) Demonstrating applications in vision, audition, realtime motor control, and sensor integration.
(e) Estimating power consumption
(f) Hypotheses testing, verification.
The Pathway to Success:-
Programmable Neuromorphic Substrate: A Digital Neurosynaptic Core
The grand challenge of neuromorphic computation is to develop a flexible brain-like architecture capable of a wide array of real-time applications, while striving towards the ultra-low power consumption and compact size of biological neural systems.It has been thought to be achieved by –
A Digital Neurosynaptic Core Using Event-Driven QDI Circuits:-
Where The innovation is a configurable neurosynaptic core that combines 256 integrate-and-fire neurons, 1024 input axons, and 1024×256 synapses in 4.2mm2 of silicon using a 45nm SOI process
TARGET ACHIEVED .
In a path to make cognitive computing systems .There was a need to do a bio metric ion of mammalian olfactory bulb .There is a system that is developed which is based on a digital neuromorphic chip consisting of 256 leaky-integrate-and-fire neurons, 1024 × 256 crossbar synapses, and address-event representation communication circuits. Thus empowering chips with a power of smell .A sense to distinguish smell and differentiate between them.
Although Cognitive Computing apart form its counter part AI has a more global algorithm . The native algorithms used in AI and genetical interpolation can be used to a certain extent in this regard. Although smell has been developed in cognitive computing system . Watson the first cognitive computing system is poised with all capabilities and is still learning .
What if the device used by the end users can touch . Using PIM(Product Information Management) as a smart Cognitive Computing database . Texture Dictionary can be developed . Although human brain as well as cognitive computing goes beyond just a simple texture classification .It can be poised with with multidimensional touch sensing and comparing as done in terms with the human brain. Imagine the situation where farmers can use cognitive computing systems to check by comparing crops just by touching .The Cognitive enabled system will compare the present status with the texture dictionary enabled with PIM . And thus will make life of farmer a lot more easy and smart .
COGNITION A SOLUTION
The basic idea is that ,as we do know that there are huge data processing problems where there is huge set of data which current computers can’t deal with and the hope is that we can develop something as form of brain which can store huge amount of data .
There is huge number of algorithms used to build chips .Each chip integrated with indiviual neurons and synapses that are built and they are connected with artitecture drawn from the brain structure.