EXACFS -- A CIL Method to mitigate Catastrophic Forgetting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Balasubramanian, S, Subramaniam, M Sai, Talasu, Sai Sriram, P, Yedu Krishna, Sai, Manepalli Pranav Phanindra, Mukkamala, Ravi, Gera, Darshan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909519236300800
author Balasubramanian, S
Subramaniam, M Sai
Talasu, Sai Sriram
P, Yedu Krishna
Sai, Manepalli Pranav Phanindra
Mukkamala, Ravi
Gera, Darshan
author_facet Balasubramanian, S
Subramaniam, M Sai
Talasu, Sai Sriram
P, Yedu Krishna
Sai, Manepalli Pranav Phanindra
Mukkamala, Ravi
Gera, Darshan
contents Deep neural networks (DNNS) excel at learning from static datasets but struggle with continual learning, where data arrives sequentially. Catastrophic forgetting, the phenomenon of forgetting previously learned knowledge, is a primary challenge. This paper introduces EXponentially Averaged Class-wise Feature Significance (EXACFS) to mitigate this issue in the class incremental learning (CIL) setting. By estimating the significance of model features for each learned class using loss gradients, gradually aging the significance through the incremental tasks and preserving the significant features through a distillation loss, EXACFS effectively balances remembering old knowledge (stability) and learning new knowledge (plasticity). Extensive experiments on CIFAR-100 and ImageNet-100 demonstrate EXACFS's superior performance in preserving stability while acquiring plasticity.
format Preprint
id arxiv_https___arxiv_org_abs_2410_23751
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle EXACFS -- A CIL Method to mitigate Catastrophic Forgetting
Balasubramanian, S
Subramaniam, M Sai
Talasu, Sai Sriram
P, Yedu Krishna
Sai, Manepalli Pranav Phanindra
Mukkamala, Ravi
Gera, Darshan
Computer Vision and Pattern Recognition
Machine Learning
Deep neural networks (DNNS) excel at learning from static datasets but struggle with continual learning, where data arrives sequentially. Catastrophic forgetting, the phenomenon of forgetting previously learned knowledge, is a primary challenge. This paper introduces EXponentially Averaged Class-wise Feature Significance (EXACFS) to mitigate this issue in the class incremental learning (CIL) setting. By estimating the significance of model features for each learned class using loss gradients, gradually aging the significance through the incremental tasks and preserving the significant features through a distillation loss, EXACFS effectively balances remembering old knowledge (stability) and learning new knowledge (plasticity). Extensive experiments on CIFAR-100 and ImageNet-100 demonstrate EXACFS's superior performance in preserving stability while acquiring plasticity.
title EXACFS -- A CIL Method to mitigate Catastrophic Forgetting
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2410.23751