YoVDO

When BERT Plays the Lottery, All Tickets Are Winning - Paper Explained

Offered By: Yannic Kilcher via YouTube

Tags

BERT Courses Machine Learning Courses Neural Networks Courses

Course Description

Overview

Explore a detailed analysis of BERT pruning in relation to the Lottery Ticket Hypothesis in this 54-minute video lecture. Delve into the fascinating discovery that even when many components of the giant BERT model are pruned away, it still maintains functionality. Learn about the experiments conducted to demonstrate that seemingly "bad" lottery tickets can be fine-tuned to achieve good accuracy. Gain insights into the reducibility of large Transformer-based models, the potential usefulness of most weights in pre-trained BERT, and the variations in "good" subnetworks across GLUE tasks. Follow along with the comprehensive outline covering BERT basics, the Lottery Ticket Hypothesis, paper abstract, pruning techniques, experimental results, and conclusions drawn from this groundbreaking research.

Syllabus

- Overview
- BERT
- Lottery Ticket Hypothesis
- Paper Abstract
- Pruning BERT
- Experiments
- Conclusion


Taught by

Yannic Kilcher

Related Courses

Introduction to Artificial Intelligence
Stanford University via Udacity
Natural Language Processing
Columbia University via Coursera
Probabilistic Graphical Models 1: Representation
Stanford University via Coursera
Computer Vision: The Fundamentals
University of California, Berkeley via Coursera
Learning from Data (Introductory Machine Learning course)
California Institute of Technology via Independent