Back

Speaker "Tsvi Achler" Details Back

 

Topic

Understanding better the information neural networks process 

Abstract

Neural networks have been greatly successful in many applications but are prone to unintuitive errors as seen with adversarial examples.
Developers struggle to design networks in better ways precisely because of difficulties obtaining intuition.
We will discuss:
1) Why it is difficult to explain neural networks.
2) How some layers process partial information and will show how developers can evaluate this information.
3) How to automatically detect potential adversarial examples that the network has not been trained on.
Our goal is to provide developers actionable insights to improve their network's performance.

Profile

Tsvi Achler has a unique background focusing on the neural mechanisms of recognition from a multidisciplinary perspective. He has done extensive work in theory and simulations, human cognitive experiments, animal neurophysiology experiments, and clinical training. He has an applied engineering background, has received bachelor degrees from UC Berkeley in Electrical Engineering, Computer Science and advanced

degrees from University of Illinois at Urbana-Champaign in Neuroscience (PhD), Medicine (MD) and worked as a postdoc in Computer Science, and at Los Alamos National Labs, and IBM Research. He now heads his own startup Optimizing Mind whose goal is to provide the

next generation of machine learning algorithms