In this episode, we interview Less Wright, who has set 13 records on the fast.ai leaderboard, and we talk about one of his tricks is using cutting-edge optimizers, he also developed one, which is called Ranger. He talks also about different strategies of using Deep Learning optimizers, it is worth to take those into consideration. We also tested Ranger and in our case, it also outperformed Adam/RAdam variants. Co-host is Levente Szabados.
Less' LinkedIn: https://www.linkedin.com/in/less-wright-22b59017/
Ranger's article: https://medium.com/@lessw/how-we-beat-the-fastai-leaderboard-score-by-19-77-a-cbb2338fab5c
Less' medium articles: https://medium.com/@lessw
His article about Mixnet he mentions: https://medium.com/@lessw/meet-mixnet-google-brains-new-state-of-the-art-mobile-ai-architecture-bd6c37abfa3a
His Github: https://github.com/lessw2020
Co-Host: Levente Szabados
https://www.linkedin.com/in/levente-szabados-76334728/
---General Info---
Podcast's Website: http://machinelearningcafe.org/
Host's LinkedIn: https://www.linkedin.com/in/miklostoth/
Email of the host: miklos@machinelearningcafe.org
---Copyright Info---
Music is from https://filmmusic.io, intro first part is by Miklos Toth and some free garage band loops. :) intro second part: "Aces High" by Kevin MacLeod, outro "Acid Trumpet" by Kevin MacLeod (https://incompetech.com), License: CC BY (http://creativecommons.org/licenses/by/4.0/)
What is beyond PoCs? ML project-hurdles you should be prepared to take with Balázs Kégl - 016
Alternative Learning Methods 01 with Dmitry Krotov - 015
"Monkeys with laptops" - the (hi)story of applying AI in real life with Denis Rothman - 014
Conversational AI at Rasa with Vladimir Vlasov - 013
What's real and what's hype? - Decades of ML with Eugene Dubossarsky - 012
Extra: Machine Learning at John Deere with Marco Lemessi - 011
The World’s first Quadruple Kaggle Grandmaster - Abhishek Thakur - 010
Everything is just a wicked graph with Benedek Rozemberczki - 009
Ratings and Marketing Attribution at Foursquare with Max Sklar - 008
MISH Activation Function with Diganta Misra - 007
Finding the label errors with Cleanlab with Curtis Northcutt - 006
Cutting-Edge Gradient Descent Variants with Levente Szabados - 004
Tree-based methods for procurement & are expert systems alive? with Renee Ahel - 003
Can we predict the accuracy of a Neural Network? Yes, with the WeightWatcher tool by Charles Martin, Ph.D. - 002
Progressively Growing GANs & Style GANs with Alexandr Honchar - 001
Create your
podcast in
minutes
It is Free
Breaking Math Podcast
Opinionated History of Mathematics
Casual Inference
The Universe Speaks in Numbers
SOA Podcasts - Society of Actuaries