Suraj Srinivas - Making Deep Neural Networks Smaller and Faster
1,285 views · Published 5 July 2016 · 17:47 · Indexed 8 November 2026
Channel: Hasgeek TV · 2016 · Science & Technology
Deep neural networks with millions of parameters are at the heart of many state of the art machine learning models today. However, is has been shown that models with much smaller number of parameters can also perform just as well. A smaller model has the advantage of being faster to evaluate and easier to store - both of which are crucial for real-time and embedded / mobile applications. In this talk, I intend to provide a brief overview of such model compression techniques. Using these techniques, it is possible to compress neural networks by as much as 10x and speed up inference by 3-4x.
More from this channel
-
3:19
Flipkart on Scaling PHP in the Cloud
-
35:00
On web standards and accessibility
-
1:05:20
Introduction to HTML5 boilerplate
-
43:45
Bits, bytes and blobs -Mrinal Wadhwa
-
52:30
A primer on ECMAScript 5 -Rajasekharan Vengalil
-
50:11
Zero to App with jQuery Mobile -Anirudh Sanjeev
-
34:26
JavaScript Gone Wild -Akash Mahajan
-
39:32
Bulding great mobile web apps: Some things you might want to know -Shwetank Dixit