INTRODUCTION:

• Neuromorphic Computing is a concept developed by

Caver Mead in late 1980’s describing the use of very-
large-scale integration (VLSI) systems containing
electronic analogue circuits to mimic neurobiological
architectures present in the nervous system.
• Neuromorphic algorithms emphasize the temporal
interaction among the processing and the memory.
Contd.,
• A neuromorphic computer is another kind of repurposable computing platform
like a CPU, GPU, FPGA, etc.
• A neuromorphic computer will be more / less efficient than another computing
architecture depending on the algorithm
• A key question in designing a neuromorphic computer is understanding the structure
of the algorithms it will likely run
• Neuromorphic computers may be good choices for implementing some machine learning
algorithms, but these should not be confused with brains
• A neuromorphic computer is not a brain, although if we were ever to figure out
how to simulate a brain on a computer, a neuromorphic computer would likely be an
efficient option.
Basic neuromorphic / cognitive computing
proposition
Build computers that learn and generalize in a broad variety of tasks, much as human
brains are able to do, in order to employ them in applications that require (too much)
human effort.

• This idea is at least 40 years old, yet we still don’t have these kinds of computers.
• We have become disillusioned with these ideas in the past because the proposition was
not fulfilled (AI and neural net “winters”)
• The proposition is (very) popular again because
• Maturation of the computing industry
• The successful application of some machine learning techniques
• Interest and research on the brain
 Neuromorphic / cognitive computing philosophy
Cognitive computing views the brain as a
Cognition = Artificial
computing
computer and thinking as the execution of
Artificial Neural
Intelligence algorithms.
Memory = Networks
(Cognitive storage of (Connectionist
Computing) data and Computing)
algorithms
Thinking = algorithm:
algorithm: application of iterative
search algorithms to
error
data
reduction
• Biological memory corresponds to a container holding data and
algorithms. Learning fills the container with input-output rules
defined on discrete (AI) or continuous (ANN) variables.
• Algorithms create input-output mappings using
rules or weights stored in memory.
• AI focuses on search algorithms to select “production” rules.
• ANN focuses on iterative error reduction algorithms to determine
“weights” yielding the desired input-output relationships.
• Algorithms are created by humans.
 NEUROMORPHIC CHIPS

• Modelled on biological brains—designed to process

sensory data such as images and sound and respond to
changes in that data in ways not specifically
programmed.
• Neuromorphic computing systems excel at computing complex
dynamics using a small set of computational primitives
(neurons, synapses, spikes).
 NEUROMORPHIC ENG'G IS NOT NEW

• A class of neuromorphic chip has been available

since 1993.
• ZISC (Zero Instruction Set Computer), became the
world’s first commercially available neuromorphic
chip.
Neuromorphic Computing Hardware Architecture
SpiNNaker (“Spiking Neural Network Architecture”)

Steve Furber, “To Build a Brain” , IEEE Spectrum, August 2012

Contd.,
• Spike
• Simplest possible temporal message
• Facilitates algorithms inspired by biological neural systems
• Supports time and rate based algorithms
• Information “packet”
• Generalization of spike time message
• A “spike” that carries additional information
• Facilitates other dynamical computing architectures using different primitives
• Routing of spikes / packets
• Messages can be packaged with an address and routed over a network (e.g. IBM, SpiNNaker)
• Messages can be delivered over a switching fabric (e.g. HRL)
• Networks can be multiscale – e.g. on core, on chip, off chip
 EXISTING CHIPS:
• IBM TrueNorth - most advanced neuromorphic (brain-like)
computer chip to date
• Qualcomm's Zeroth Program
• Human Brain Project in Europe
• Nvidia's Tegra X1 revealed at CES 2015
• Intel’s Loihi Chip.
EXISTING APPLICATION
 MORAVEC'S PARADOX

• Sensory information processing is extremely easy for brains

but extremely hard for modern computers, whereas
symbolic information processing is comparably hard for
brains but extremely easy for modern computers.
 Future Applications

• Image based Web-searching

Contd.,

• Advancements in Facial/Biometric Recognition:

ATTENDANCE IN A SNAP
 Contd.,
• Development of Smart Medical Devices.

AUTO-ECG
 Advantages
• Can compute in real time, which is similar to how
the brain works
• Might take us a step closer to artificial intelligence
• Different applications in various disciplines
• Advances in neuroscience and chip making
Key Technology Issues / Choices
• Distributing large amounts of memory (synapses) among many
processors (neurons) on a single chip.
• Developing a scalable messaging (spiking) architecture.
• Selection of computational primitives (e.g. neuron and synapse
models)
• Engineering for scale, space and power efficiency
• Creating a large-scale simulation capability that accurately
models the neuromorphic hardware
• Creating tools to develop and debug neural algorithms on the
simulator and
• the neuromorphic hardware
• Writing the algorithms (including those that learn)
Key Challenges in Building a Brain
1. Brains are not Computers.
2. Computers can compute anything.
3. Competition for Resources
4. Massive Computing Resources
5. No Easy Path for Technology Evolution
6. Whole System Interdependence
7. Whole System Requirement
 Contd.,
1.Brains are not Computers:
• Brains are thermodynamical, bio/chemo/ physical systems that evolved from and are
embedded in the natural world
• Computers are symbolic processors executing algorithms designed by humans
• Brains designed computers.

2 . Computers can compute anything:

• The computer is a blank slate
• We must generate all the constraints to build a neuromorphic computer
• Changing computing architecture only changes the classes of algorithms that
• it computes efficiently
Contd.,
3. Competition for Resources:
• It is easy for anyone who doesn’t like your project to claim that
• It is making no progress
• It is not competitive with the state of the art
• You are doing it wrong
• You are an idiot
4. Massive Computing Resources:
• Any model that does anything that anyone will care about requires a
massive computational resource for development and implementation
• Development is slow and expensive
• Custom hardware in state of art process is needed for any large
scale application
Contd.,
5. No Easy Path for Technology Evolution:
• The benchmark for performance comparison is either
• A human
• A well-engineered, domain-specific solution

6. Whole System Interdependence:

• Brains / bodies / environments are complex systems whose large scale
function (almost certainly) cannot be analytically expressed in terms of its
lower level structure / dynamics
• System design methodologies are inadequate because the system cannot be
decomposed into independent parts
Contd.,

7. Whole System Requirement:

• Brains are embodied and bodies are embedded in an
environment (Edelman)
• Testing often requires embedding the neuromorphic computer
in a complex
• body /environment.
 References
• Al-Rodhan, N. (2015, August 14). The Moral Code. Retrieved September 11,
2015, from https://www.foreignaffairs.com/articles/2015-08-12/moral-code
• Al-Rodhan, N. (2015, March 13). The Many Ethical Implications of Emerging Technologies. Retrieved
September 13, 2015, from http://www.scientificamerican.com/article/the-many-ethical-implications-of-
emerging-technologies/
• Easton. (2015, March 13). Top 5 Emerging Technologies In 2015. Retrieved September 13, 2015, from
• https://wtvox.com/robotics/top-5-emerging-technologies-in-2015/
• Gershgorn, D. (2015, August 19). IBM Chips Could Make Your Smartphone Think Like A Mouse. Retrieved
September 11, 2015, from http://www.popsci.com/ibms-neuromorphic-chip-array-mimics-rodent-brain
• Hof, R. (2014, April 23). Qualcomm's Neuromorphic Chips Could Make Robots and Phones More Astute
About the World. Retrieved September 9, 2015, from
http://www.technologyreview.com/featuredstory/526506/neuromorphic- chips/
• Lambinet, P. (2015, January 31). The Ongoing Quest For The ‘Brain’ Chip. Retrieved September 12,
2015, from http://techcrunch.com/2015/01/31/the-ongoing-quest-for-the-brain-chip/
• Lovetrue, B. (2015, May 15). 4 Security Challenges That May Tear Apart the Internet of Things | EE Times.
Retrieved October 18, 2015, from
http://www.eetimes.com/author.asp?section_id=36&doc_id=1326549&page_number=2
Contd.,
• Metz, C. (2015, August 17). IBM’s ‘Rodent Brain’ Chip Could Make Our Phones Hyper-Smart. Retrieved
• September 11, 2015, from http://www.wired.com/2015/08/ibms-rodent-brain-chip-make-phones-hyper-
smart/
• Moore-Colyer, R. (2015, August 27). Apple, Google, Microsoft and Nvidia race ahead with car tech plans.
• Retrieved September 11, 2015, from http://www.theinquirer.net/inquirer/feature/2423008/apple-google-
• microsoft-nvidia-and-more-accelerate-car-tech-plans
• Poeter, D. (2013, October 14). Qualcomm Demos Brain-Inspired Zeroth Chips. Retrieved September 9,
2015, from http://www.pcmag.com/article2/0,2817,2425743,00.asp
• http://www.research.ibm.com/cognitive-computing/neurosynaptic-chips.shtml#fbid=Wnq8mAzUKSc
• Tarantola, A. (2015, August 17). IBM wires up 'neuromorphic' chips like a rodent's brain. Retrieved
• September 11, 2015, from http://www.engadget.com/2015/08/17/ibm-wires-up-neuromorphic-chips-
• like-a-rodents-brain/
• Top 10 Emerging Technologies of 2015. (2015, March). Retrieved September 13, 2015, from
http://www.weforum.org/reports/top-10-emerging-technologies-2015
THANK YOU