paper-with-me

홈 › Papers

Generative replay with feedback connections as a general strategy for continual learning

2018-09-27 · Gido M. van de Ven, Andreas S. Tolias

A major obstacle to developing artificial intelligence applications capable of true lifelong learning is that artificial neural networks quickly or catastrophically forget previously learned tasks when trained on a new one. Numerous methods for alleviating catastrophic forgetting are currently being proposed, but differences in evaluation protocols make it difficult to directly compare their performance. To enable more meaningful comparisons, here we identified three distinct scenarios for continual learning based on whether task identity is known and, if it is not, whether it needs to be inferred. Performing the split and permuted MNIST task protocols according to each of these scenarios, we found that regularization-based approaches (e.g., elastic weight consolidation) failed when task identity needed to be inferred. In contrast, generative replay combined with distillation (i.e., using class probabilities as "soft targets") achieved superior performance in all three scenarios. Addressing the issue of efficiency, we reduced the computational cost of generative replay by integrating the generative model into the main model by equipping it with generative feedback or backward connections. This Replay-through-Feedback approach substantially shortened training time with no or negligible loss in performance. We believe this to be an important first step towards making the powerful technique of generative replay scalable to real-world continual learning applications.

📄 PDF Abstract BibTeX arXiv:1809.10635

Code (5)

GMvandeVen/continual-learning 공식 구현 pytorch
WangTianduo/lifelong-learning-3-cases pytorch
XSMUBC/DNC-lifelong-learning pytorch
XSMUBC/Lifelong-learning_xsm pytorch
llzlcl/Continual-Learning pytorch

Tasks

Continual LearningLifelong learningPermuted-MNIST

Similar Papers 제목 키워드 기반

Brain-inspired replay for continual learning with artificial neural networks

2020-08-13 · Gido M. van de Ven, Hava T. Siegelmann & Andreas S. Tolias

Artificial neural networks suffer from catastrophic forgetting. Unlike humans, when these networks are trained on something new, they rapidly forget what was learned before. In the brain, a mechanism thought to be import…

class-incremental learningClass Incremental LearningContinual LearningIncremental Learning

Generative Feature Replay with Orthogonal Weight Modification for Continual Learning

2020-05-07 · Gehui Shen, Song Zhang, Xiang Chen, Zhi-Hong Deng

The ability of intelligent agents to learn and remember multiple tasks sequentially is crucial to achieving artificial general intelligence. Many continual learning (CL) methods have been proposed to overcome catastrophi…

class-incremental learningClass Incremental LearningContinual LearningIncremental Learning

Replay to Remember (R2R): An Efficient Uncertainty-driven Unsupervised Continual Learning Framework Using Generative Replay

2025-05-07 · Sriram Mandalika, Harsha Vardhan, Athira Nambiar

Continual Learning entails progressively acquiring knowledge from new data while retaining previously acquired knowledge, thereby mitigating ``Catastrophic Forgetting'' in neural networks. Our work presents a novel uncer…

Continual Learning

When Generative Replay Meets Evolving Deepfakes: Domain-Aware Relative Weighting for Incremental Face Forgery Detection

2025-11-23 · Hao Shen, Jikang Cheng, Renye Yan, Zhongyuan Wang 외 arxiv

The rapid advancement of face generation techniques has led to a growing variety of forgery methods. Incremental forgery detection aims to gradually update existing models with new forgery data, yet current sample replay…

Incremental Learning

Class-Incremental Learning Using Generative Experience Replay Based on Time-aware Regularization

2023-10-05 · Zizhao Hu, Mohammad Rostami

Learning new tasks accumulatively without forgetting remains a critical challenge in continual learning. Generative experience replay addresses this challenge by synthesizing pseudo-data points for past learned tasks and…

class-incremental learningClass Incremental LearningContinual LearningIncremental Learning