Neuromorphic computing

An introduction into Neuromorphic Computing, through Why, What, How and When

1. Neuromorphic Computing - An intro to building brains

2. What is Neuromorphic computing? "Neuromorphic engineering, also known as neuromorphic computing started as a concept developed by Carver Mead in the late 1980s, describing the use of very-large-scale integration (VLSI) systems containing electronic analogue circuits to mimic neurobiological architectures present in the nervous system."

3. simply put... "Neuromorphic engineering is a new emerging interdisciplinary field which takes inspiration from biology, physics, mathematics, computer science and engineering to design hardware/physical models of neural and sensory systems."

4. Moore's Law In 1965, Gordon Moore made a prediction that would set the pace for our modern digital revolution. From careful observation of an emerging trend, Moore extrapolated that computing would dramatically increase in power, and decrease in relative cost, at an exponential pace. The insight, known as Moore’s Law, became the golden rule for the electronics industry, and a springboard for innovation

5. Analogy Biology Machine learning

6. Analogy Biology Machine learning Arbor Connectivity

7. Analogies Biology Neuron Arbor Scale 1. ~100B Neurons 2. ~100T connections Machine learning Neuron/Filter/Feature extractor Connectivity Scale ???

8. Analogies Biology Neuron Arbor Power ~20W Machine learning Neuron/Filter/Feature extractor Connectivity Power ???

9. Neuromorphic Computing Neurogrid IBM TrueNorth

10. Neurogrid 1 16 chip system, which emulates a million neurons with a billion connection 2 Morphs analog property of neurons of brain by using sub threshold analog logic 3 Extremely low level transistor circuit that could emulate the non linearities that are captured by neurons of brain 4 Asynchronous digital logic for communication

11. TrueNorth 1 Comes from IBM's cognitive computing division 2 16x times the size of Neurogrid 3 Instead of subthreshold analog, they are completely digital

12. Reference 1 Ben Varkey Benjamin ; Peiran Gao ; Emmett McQuinn ; Swadesh Choudhary ; Anand R. Chandrasekaran ; Jean-Marie Bussat ; Rodrigo Alvarez-Icaza ; John V. Arthur ; Paul A. Merolla ; Kwabena Boahen - Neurogrid: A Mixed-Analog-Digital Multichip System for Large-Scale Neural Simulations - Stanford University 2 Wen Ma: Mohammed A. Zidan - Neuromorphic computing with memristive devices - Springer journal

13. Thank you