Priyadarshini Panda

PhD Student, Purdue University

ABOUT          RESEARCH           PUBLICATIONS          RESUME


Text Box:

  

I am a PhD student at Purdue University, West Lafayette, USA working with Prof. Kaushik Roy in Nanoelectronics Research Lab. My research interests lie in Neuromorphic Computing, specifically developing scalable energy-efficient design methodologies for deep learning applications (recognition, inference, analytics), novel supervised/unsupervised learning algorithms for feedforward spiking/dynamic reservoir networks for spatio-temporal data processing, developing robust learning frameworks for tasks such as reinforcement learning, sequence generation etc. Presently, in the spiking domain, I have been working on developing spike based learning algorithms capable of continuous/ lifelong learning in a dynamic environment. On the other hand, in deep learning domain, I have been studying the robustness of state of the art deep nets toward adversarial attacks utilizing Principal Component Analysis. I am excited by the possibility of unravelling the black-box of artificial learning using statistical concepts from random matrix theory such as Eigen Value Spectra, Mean Field analytics and Linear Approximation Theory.

 

 

 

 

 

 

 

 

 

“Can machines think?” , the question brought up by Alan Turing, has led to the development of the field of brain-inspired computing wherein researchers have put substantial effort in building smarter devices and technology that have the potential of human-like understanding. However, there still remains a large (several orders-of-magnitude) power efficiency gap between the human brain and computers that attempt to emulate even some facets of its functionality. Today, learning efficiently and appropriately are key to achieving a functional cognitive system. While deep learning has no doubt paved the way to learning appropriately, the energy efficiency and compute power associated with learning still remains a concern. In this regard, the GOAL of my research is to harness efficiency in cognitive applications using novel algorithms, architectures and design techniques. Please see my RESEARCH page for more details.

 

 

 

 

 

 

 

 

 

 

 

 

 

pandap@purdue.edu

NRL                  Purdue University                   ECE

Copyright © 2018 Priyadarshini Panda