This video is based on the following series of lectures:
Learning from data. Professor Yaser Abu-Mostafa, Caltech: https://tinyurl.com/4wkr7prx
Machine learning Theory. Professor Shai Ben-David and Shai Shalev-Shwartz: https://tinyurl.com/26v5btve
In this video we take a look at the strict Statistical Learning Theory framework for Supervised Classification. We take a quick look at Hoeffding's inequality, PAC Learning, the bias-complexity tradeoff and the feasibility of learning.
Timestamps :
0:00 – Intro
0:15 – Law of Large numbers
2:47 – Hoeffding’s Inequality
5:46 – Feasibility of Learning for Finite Hypothesis Classes
10:32 –The bias-complexity tradeoff
11:18 – Need for a better measure of complexity?
11:46 – The same is true for stochastic distributions as well!
_______________________________________
References:
Books:
Understanding Machine Learning: From Theory to Algorithms | Shai Shalev-Shwartz and Shai Ben-David
Learning From Data | Yaser Abu-Mostafa ,Malik Magdon-Ismail , Hsuan-Tien Lin
Foundations of Machine Learning | Mehryar Mohri, Afshin Rostamizadeh, and Ameet Talwalkar
__________________________________________________
Social Media:
Kartik Chincholikar Website: https://kartikchincholikar.github.io/
Github: https://github.com/kartikchincholikar
Twitter: https://twitter.com/KartikC14
_______________________________________________
I love coffee! Help fund future projects: https://www.buymeacoffee.com/karti
_______________________________________________
Music:
Track: Such Memories
Music composed and recorded by Oak Studios
Creative Commons - Attribution ND 4.0 https://youtu.be/GspVwN-9_Fs