Machine Learning Paper Discussion: BackDrop and Stochastic Delta Rule

Hey Everyone,

Our last discussion was a lot of fun so we are doing it again! Just like last time, we will be located at the zesty.ai offices near Jack London Square.

This time, we'll be diving into some exciting Dropout variants with the following papers:

Backdrop: Stochastic Backpropagation
(https://arxiv.org/pdf/1806.01337.pdf) - Siavash Golkar and Kyle Cranmer

Dropout is a special case of the stochastic delta rule: faster and more accurate deep learning
(https://arxiv.org/pdf/1808.03578.pdf) - Noah Frazier-Logue and Stephen José Hanson

The meetup will begin with a presentation of the key points of the paper, followed by a group discussion. Reading is not strictly required, but it will be necessary to contribute to the discussion.

We look forward to seeing you all there! Please only RSVP if you plan on attending!

HOW TO FIND US:

Entrance is at 331 Jefferson St. Please knock on the door or ring the doorbell and someone will let you in.

DRIVING: If you will be driving, there is plenty of street parking near the building.

PUBLIC TRANSIT: The closest BART stop is 12th street station. We're also near the ferry in Jack London Square.