Machine Learning Yearning

Sigo publicando Machine Learning Yearning:

AI is the new electricity

Machine Learning Yearning Dear friends, Last week, we discussed what happens when your training set and dev/test set have different data distributions. How can you tell if your algorithm’s performance is not generalizing well to a different distribution than what it was trained on? We call this problem data mismatch. This week, you’ll how learn to diagnose data mismatch, as opposed to bias and variance. Once you’ve identified it, you will also learn how to address data mismatch with techniques such as artificial data synthesis. Read on to learn more! Read Chapters 40-43 In Case You Missed It: Stanford CS230 (Deep Learning) Poster Session Stanford’s CS230 class had a total of ~800 students this past academic year (including 350 in the most recent offering). I think this is the fastest a class has grown from 0 to 800 students in Stanford history! Saturday’s projects included applications from disease diagnosis to bitcoin prediction to a Pokémon HP predictor. If you are interested in reading about these student projects, we will be posting the posters online soon and will let you know when they’re up. Pun of the Week If you have a great AI pun, tweet it to me @AndrewYNg using #AIpun. I'll share my favorite in next week's email!

Catch up on Machine Learning Yearning Miss last week's email? Access all released chapters Copyright © 2018 Andrew Ng, All rights reserved. You are receiving this because you opted in to receiving emails about Andrew's upcoming book. Our mailing address is: Andrew Ng 353 Serra Mall Stanford, CA 94305 Add us to your address book Want to change how you receive these emails? You can update your preferences or unsubscribe from this list.