Beyond Not-Forgetting: Continual Learning with Backward Knowledge Transfer
Nov 28, 2022
Speakers
Deliang Fan
Speaker · 0 followers
Junshan Zhang
Speaker · 0 followers
About
By learning a sequence of tasks continually, an agent in continual learning (CL) can improve the learning performance of both a new task and `old' tasks by leveraging the forward knowledge transfer and the backward knowledge transfer, respectively. However, most existing CL methods focus on addressing catastrophic forgetting in neural networks by minimizing the modification of the learnt model for old tasks. This inevitably limits the backward knowledge transfer from the new task to the old tasks, because judicious model updates could possibly improve the learning performance of the old tasks as well. To tackle this problem, we first theoretically analyze the conditions under which updating the learnt model of old tasks could be beneficial for CL and also lead to backward knowledge transfer, based on the gradient projection onto the input subspaces of old tasks. Building on the theoretical analysis, we next develop a ContinUal learning method with Backward knowlEdge tRansfer (CUBER), for a fixed capacity neural network without data replay. In particular, CUBER first characterizes the task correlation to identify the positively correlated old tasks in a layer-wise manner, and then selectively modifies the learnt model of the old tasks when learning the new task. Experimental studies show that CUBER can even achieve positive backward knowledge transfer on several existing CL benchmarks for the first time without data replay, where the related baselines still suffer from catastrophic forgetting (negative backward knowledge transfer). The superior performance of CUBER on the backward knowledge transfer also leads to higher accuracy accordingly.By learning a sequence of tasks continually, an agent in continual learning (CL) can improve the learning performance of both a new task and `old' tasks by leveraging the forward knowledge transfer and the backward knowledge transfer, respectively. However, most existing CL methods focus on addressing catastrophic forgetting in neural networks by minimizing the modification of the learnt model for old tasks. This inevitably limits the backward knowledge transfer from the new task to the old task…
Organizer
NeurIPS 2022
Account · 952 followers
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
InterFair: Debiasing with Natural Language Feedback for Fair Interpretable Predictions
Watch later
Bodhisattwa Prasad Majumder, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Featured Papers Panels 2A
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Towards Algorithmic Fairness in Space-Time: Filling in Black Holes
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Just Avoid Robust Inaccuracy: Boosting Robustness Without Sacrificing Accuracy
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Computationally Efficient Horizon-Free Reinforcement Learning for Linear Mixture MDPs
Watch later
Dongruo Zhou, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
GET3D: A Generative Model of High Quality 3D Textured Shapes Learned from Images
Watch later
Jun Gao, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%