Baby Got Backpropagation

Location

CoLab, OCB 100

Start Date

27-4-2018 9:00 AM

Document Type

Poster

Description

Backpropagation is the algorithm for determining how a single training example would nudge the weights and biases, not just in terms of whether they should go up and down, but in terms of what relative proportion to those changes find the most rapid decrease to the cost. Because of this, it has become the most popular algorithm for machine learning programs.

Comments

The faculty supervisor for this project is Lauren Jacobs, Mathematics.

Image

This document is currently not available here.

Share

COinS
 
Apr 27th, 9:00 AM

Baby Got Backpropagation

CoLab, OCB 100

Backpropagation is the algorithm for determining how a single training example would nudge the weights and biases, not just in terms of whether they should go up and down, but in terms of what relative proportion to those changes find the most rapid decrease to the cost. Because of this, it has become the most popular algorithm for machine learning programs.