Machine Learning / Data Science Reading Group: Why Adam Outperforms Gradient Descent
Details
The next reading is: Heavy-Tailed Class Imbalance and Why Adam Outperforms Gradient Descent on Language Models
This meetup occurs online and in-person.
Please try to read it before the meetup. It's OK if you don't understand it all. The meetup is for questions and discussion.
Important!: To attend this event in-person, you must fill out our registration form with your name and email address by noon the day before the event. This is a requirement of our hosts at Northeastern University. Building security will not let in people whose names do not appear on the registration list. RSVPing through this event page is not enough. Your registration information will be shared with event organizers, staff at the venue, and third-party tools used by the venue to manage visitors.
Microsoft Teams Link will be provided on the day.
Schedule:
• 6:00PM Virtual doors are open, feel free to mingle
• 6:30PM Reading group starts
***
What is the Data Science Reading Group about?:
We will meet every other week and present/discuss a topic of interest.
The group votes on the topic/paper to read next time. Anyone can propose a paper for the following Meetup.
Discussions about this meetup also happen on Discord. Go here to join the #data-science channel.
A list of past papers, both discussed and proposed, is available here: https://github.com/learn-data-science/data-science-reading