A Theoretical Analysis of Catastrophic Forgetting through the NTK Overlap Matrix

10/07/2020
by   Thang Doan, et al.
0

Continual learning (CL) is a setting in which an agent has to learn from an incoming stream of data during its entire lifetime. Although major advances have been made in the field, one recurring problem which remains unsolved is that of Catastrophic Forgetting (CF). While the issue has been extensively studied empirically, little attention has been paid from a theoretical angle. In this paper, we show that the impact of CF increases as two tasks increasingly align. We introduce a measure of task similarity called the NTK overlap matrix which is at the core of CF. We analyze common projected gradient algorithms and demonstrate how they mitigate forgetting. Then, we propose a variant of Orthogonal Gradient Descent (OGD) which leverages structure of the data through Principal Component Analysis (PCA). Experiments support our theoretical findings and show how our method reduces CF on classical CL datasets.

READ FULL TEXT

page 17

page 18

page 19

page 20

page 21

page 26

research
06/21/2020

Generalisation Guarantees for Continual Learning with Orthogonal Gradient Descent

In continual learning settings, deep neural networks are prone to catast...
research
02/07/2023

Utility-based Perturbed Gradient Descent: An Optimizer for Continual Learning

Modern representation learning methods may fail to adapt quickly under n...
research
04/26/2022

Theoretical Understanding of the Information Flow on Continual Learning Performance

Continual learning (CL) is a setting in which an agent has to learn from...
research
05/16/2021

Statistical Mechanical Analysis of Catastrophic Forgetting in Continual Learning with Teacher and Student Networks

When a computational system continuously learns from an ever-changing en...
research
02/12/2023

Theory on Forgetting and Generalization of Continual Learning

Continual learning (CL), which aims to learn a sequence of tasks, has at...
research
01/23/2020

Structured Compression and Sharing of Representational Space for Continual Learning

Humans are skilled at learning adaptively and efficiently throughout the...
research
10/15/2019

Embodiment dictates learnability in neural controllers

Catastrophic forgetting continues to severely restrict the learnability ...

Please sign up or login with your details

Forgot password? Click here to reset