[1606.09282] Learning without Forgetting
Abstract:When building a unified vision system or gradually adding new capabilities to a system, the usual assumption is that training data for all tasks is always available. However, as the number of tasks grows, storing and retraining on such data becomes infeasible. A new problem arises where we add new capabilities to a Convolutional Neural Network (CNN), but the training data for its existing capabilities are unavailable. We propose our Learning without Forgetting method, which uses only new task data to train the network while preserving the original capabilities. Our method performs favorably compared to commonly used feature extraction and fine-tuning adaption techniques and performs similarly to multitask learning that uses original task data we assume unavailable. A more surprising observation is that Learning without Forgetting may be able to replace fine-tuning with similar old and new task datasets for improved new task performance.
1 Learning without Forgetting Zhizhong Li, Derek Hoiem, Member, IEEE Abstract—When building a unified vision system or gradually adding new capabilities to a system, the usual assumption is that training data for all tasks is always available. However, as the number of tasks grows, storing and retraining on such data becomes infeasible. A…
saved by
related reading
- Learning without Forgettingarxiv.org
- [2605.15220] Always Learning, Always Mixing: Efficient and Simple Data Mixing All The Timearxiv.org
- [2605.15220] Always Learning, Always Mixing: Efficient and Simple Data Mixing All The Timearxiv.org
- The Continual Learning Problemjessylin.com
- Self-Distillation Enables Continual Learningarxiv.org
- [1706.04599] On Calibration of Modern Neural Networksarxiv.org
- Understanding deep learning requires rethinking generalizationarxiv.org
- Feature-wise transformationsdistill.pub
- [1810.13128] The Effect of Learning Strategy versus Inherent Architecture Properties on the Ability of Convolutional Neural Networks to Develop Transformation Invariancearxiv.org
- ImageNet Classification with Deep Convolutional Neural Networksproceedings.neurips.cc
- 1011.0686arxiv.org
- Learning to Remember — Lawrence Fengar-forum.github.io