paper-with-me

홈 › Papers

Overcoming catastrophic forgetting with hard attention to the task

2018-01-04 · ICML 2018 7 · Joan Serrà, Dídac Surís, Marius Miron, Alexandros Karatzoglou

Catastrophic forgetting occurs when a neural network loses the information learned in a previous task after training on subsequent tasks. This problem remains a hurdle for artificial intelligence systems with sequential learning capabilities. In this paper, we propose a task-based hard attention mechanism that preserves previous tasks' information without affecting the current task's learning. A hard attention mask is learned concurrently to every task, through stochastic gradient descent, and previous masks are exploited to condition such learning. We show that the proposed mechanism is effective for reducing catastrophic forgetting, cutting current rates by 45 to 80%. We also show that it is robust to different hyperparameter choices, and that it offers a number of monitoring capabilities. The approach features the possibility to control both the stability and compactness of the learned knowledge, which we believe makes it also attractive for online learning or network compression applications.

📄 PDF Abstract BibTeX arXiv:1801.01423

Code (2)

joansj/hat 공식 구현 pytorch
chilung/hat pytorch

Tasks

Continual LearningHard Attention

Similar Papers 제목 키워드 기반

Attention-Based Structural-Plasticity

2019-03-02 · Soheil Kolouri, Nicholas Ketz, Xinyun Zou, Jeffrey Krichmar 외

Catastrophic forgetting/interference is a critical problem for lifelong learning machines, which impedes the agents from maintaining their previously learned knowledge while learning new tasks. Neural networks, in partic…

Lifelong learningPermuted-MNISTSplit-MNIST

ZeroFlow: Overcoming Catastrophic Forgetting is Easier than You Think

2025-01-02 · Tao Feng, Wei Li, Didi Zhu, Hangjie Yuan 외

Backpropagation provides a generalized configuration for overcoming catastrophic forgetting. Like, SGD and Adam are commonly used for weight updates in continual learning and continual pre-training. In practice, permissi…

Continual Learning

Online Structured Laplace Approximations For Overcoming Catastrophic Forgetting

2018-05-20 · NeurIPS 2018 12 · Hippolyt Ritter, Aleksandar Botev, David Barber

We introduce the Kronecker factored online Laplace approximation for overcoming catastrophic forgetting in neural networks. The method is grounded in a Bayesian online learning framework, where we recursively approximate…

Permuted-MNIST

Overcoming Catastrophic Forgetting beyond Continual Learning: Balanced Training for Neural Machine Translation

2022-03-08 · ACL 2022 5 · Chenze Shao, Yang Feng

Neural networks tend to gradually forget the previously learned knowledge when learning multiple tasks sequentially from dynamic data distributions. This problem is called \textit{catastrophic forgetting}, which is a fun…

Continual LearningKnowledge DistillationMachine TranslationTranslation

Overcoming Catastrophic Forgetting in Graph Neural Networks

2020-12-10 · Huihui Liu, Yiding Yang, Xinchao Wang

Catastrophic forgetting refers to the tendency that a neural network "forgets" the previous learned knowledge upon learning new tasks. Prior methods have been focused on overcoming this problem on convolutional neural ne…

Continual Learning