Progressive Learning without Forgetting

11/28/2022
by   Tao Feng, et al.
0

Learning from changing tasks and sequential experience without forgetting the obtained knowledge is a challenging problem for artificial neural networks. In this work, we focus on two challenging problems in the paradigm of Continual Learning (CL) without involving any old data: (i) the accumulation of catastrophic forgetting caused by the gradually fading knowledge space from which the model learns the previous knowledge; (ii) the uncontrolled tug-of-war dynamics to balance the stability and plasticity during the learning of new tasks. In order to tackle these problems, we present Progressive Learning without Forgetting (PLwF) and a credit assignment regime in the optimizer. PLwF densely introduces model functions from previous tasks to construct a knowledge space such that it contains the most reliable knowledge on each task and the distribution information of different tasks, while credit assignment controls the tug-of-war dynamics by removing gradient conflict through projection. Extensive ablative experiments demonstrate the effectiveness of PLwF and credit assignment. In comparison with other CL methods, we report notably better results even without relying on any raw data.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
11/16/2020

Gradient Episodic Memory with a Soft Constraint for Continual Learning

Catastrophic forgetting in continual learning is a common destructive ph...
research
10/11/2021

Addressing the Stability-Plasticity Dilemma via Knowledge-Aware Continual Learning

Continual learning agents should incrementally learn a sequence of tasks...
research
06/19/2020

SOLA: Continual Learning with Second-Order Loss Approximation

Neural networks have achieved remarkable success in many cognitive tasks...
research
06/09/2021

Optimizing Reusable Knowledge for Continual Learning via Metalearning

When learning tasks over time, artificial neural networks suffer from a ...
research
11/29/2022

Lifelong Person Re-Identification via Knowledge Refreshing and Consolidation

Lifelong person re-identification (LReID) is in significant demand for r...
research
02/02/2023

Continual Learning with Scaled Gradient Projection

In neural networks, continual learning results in gradient interference ...
research
04/27/2020

A general approach to progressive intelligence

In biological learning, data is used to improve performance on the task ...

Please sign up or login with your details

Forgot password? Click here to reset