Phase Change Memory Devices for Cognitive Computing
Mathematical modeling of neural circuits plays a key role in understanding the fundamental mechanisms underlying brain function. However, simulation of large-scale neural circuits on digital computers is challenging, particularly in terms of speed of simulation and power efficiency. Thanks to recent developments in nanotechnology, special purpose neuromorphic hardware is emerging to be a more energy efficient alternative for this task. This has also buoyed the hopes for quickly translating neuro-scientific discoveries into portable cognitive computing technologies that interact with the environment in an intelligent manner.
In this talk, I will describe our results on the development of nanoscale, low power synaptronic devices based on Phase Change Memory (PCM) technology. I will show that these devices can be programmed to mimic the plasticity of biological synapses, approaching the energy efficiency of the human brain. I will also show that they support efficient learning of simple temporal sequences by spiking neural networks.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
- 161 Warren Street
- Newark, New Jersey
- United States 07102
- Building: ECE Building, Room 202, NJIT
- Room Number: ECEC-201
- Click here for Map
- Contact Event Host
- Co-sponsored by NJIT
- Starts 08 February 2012 05:00 PM UTC
- Ends 23 February 2012 10:00 PM UTC
- No Admission Charge
Speakers
Bipin Rajendran of IBM T. J. Watson Research Center
Phase Change Memory Devices for Cognitive Computing
Mathematical modeling of neural circuits plays a key role in understanding the fundamental mechanisms underlying brain function. However, simulation of large-scale neural circuits on digital computers is challenging, particularly in terms of speed of simulation and power efficiency. Thanks to recent developments in nanotechnology, special purpose neuromorphic hardware is emerging to be a more energy efficient alternative for this task. This has also buoyed the hopes for quickly translating neuro-scientific discoveries into portable cognitive computing technologies that interact with the environment in an intelligent manner. In this talk, I will describe our results on the development of nanoscale, low power synaptronic devices based on Phase Change Memory (PCM) technology. I will show that these devices can be programmed to mimic the plasticity of biological synapses, approaching the energy efficiency of the human brain. I will also show that they support efficient learning of simple temporal sequences by spiking neural networks.
Biography: Dr. Bipin Rajendran is a Master Inventor and Research Staff Member at IBM T. J. Watson Research Center, engaged in exploratory research on non-volatile memories and neuromorphic computation. He has published more than 30 papers in peer reviewed journals and conferences, and has been issued 20 US patents. He received a B.Tech degree (2000) from Indian Institute of Technology, Kharagpur and M.S (2003) and Ph.D (2006) in Electrical Engineering from Stanford University.
Email:
Address:Thomas J. Watson Research Center, , Yorktown Heights, New York, United States
Bipin Rajendran of IBM T. J. Watson Research Center
Phase Change Memory Devices for Cognitive Computing
Biography:
Email:
Address:Yorktown Heights, New York, United States
Agenda
Time: 5:00 PM, Thursday, February 23, 2012. Refreshments (Pizza & Soda) will be at 4:45 PM.