Fast, Portable and Accurate Neural Networks

Algorithmic optimization techniques to improve network performance, including network compression techniques such as network pruning, low precision, and sparsity. Effective techniques to improve inference speed and portability, including a comparison of the classic approach vs. a more accurate approach to considering flops, parameters, compute, memory, and heap size.


Outline/Structure of the Talk


Optimizing the performance of Neural Nets

Making Networks Portable without losing Accuracy

Learning Outcome

How to optimize the networks without compromising on accuracy.

Target Audience

Ai Developers, Data Scientists, AI Engineers

Prerequisites for Attendees

Thorough knowledge in Deep Learning mainly Neural Networks and Computer Vision.

schedule Submitted 1 year ago

Public Feedback

    • Sri Harsha Gajavalli

      Sri Harsha Gajavalli - Interpretable AI for Network Security - IIDS

      Sri Harsha Gajavalli
      Sri Harsha Gajavalli
      schedule 1 year ago
      Sold Out!
      20 Mins

      In today's world, Network and System Security are of paramount importance in the digital communication environment. On par with the developments in technology, many threats have emerged for information security which has worse effects when it comes to sensitive transactions.

      Intruders can easily break the walls of the network and can cause many kinds of breaches such as the crash of the networks, Denial Of Service, injecting Malware, etc.,

      An Intelligent Intrusion Detection System is built by leveraging the benefits of Machine Learning techniques that detect the intrusion and alerts the respective network administrator. The developed system analyses and predicts the behavior of users which in turn classifies as an anomaly or normal behavior.