SBASSE Biology Talk-2016
Computation in Neural Systems: From Biophysics to Deep Learning, Neuromorphic Engineering and beyond
Guest Speaker:
Asim Iqbal
Date: 20th December, 2016 (Tuesday)
Time: 3:0 pm – 4:0 pm
Venue: Smart Room, SBASSE 5th Floor
Host: Dr. Muhammad Tariq
Abstract:
Human brain is a fascinating device in which a single neuron is able to remarkably integrate the spiking information coming from a network of hundreds, thousands and millions of neurons. Taking this inspiration from neurobiology, we are able to model a large scale deep neural network which is currently state-of-the-art in real-world applications such as object classification, speech recognition, etc. To further improve the performance of Artificial Intelligent Systems, a deep neural network would only run best on a brain-inspired hardware. In this perspective, neuromorphic engineering stages a playground to emulate the computational functionality of neurons onto a single electronic chip.
In this talk, the functionality of a neuron, deep neural networks and how do we implement these models on a neuromorphic hardware will be presented. Furthermore, I will be scratching the surface about the impact of computational neuroscience and deep learning in both academics and industry.
Biography:
Asim completed his Masters in Neural Systems and Computation (2015) at UZH/ETH Zurich and currently pursuing his PhD in the same field. He has worked as a researcher in the Brain-inspired computing team at IBM Research-Almaden, CA and Brain and Cognitive Sciences department at Massachusetts Institute of Technology (MIT), Cambridge, MA. His research work is focused on neural circuits, deep neural networks and neuromorphic engineering.