Continual Learning via Bit-Level Information Preserving

By Yujun Shi, Li Yuan, Yunpeng Chen, Jiashi Feng. 2021

54 citations recorded by the source corpus.

Continual learning tackles the setting of learning different tasks sequentially. Despite the lots of previous solutions, most of them still suffer significant forgetting or expensive memory cost. In this work, targeted at these problems, we first study the continual learning process through the lens of information theory and observe that forgetting of a model stems from the loss of information gain on its parameters from the previous tasks when learning a new task. From this viewpoint, we then propose a novel continual learning approach called Bit-Level Information Preserving (BLIP) that preserves the information gain on model parameters through updating the parameters at the bit level, which can be conveniently implemented with parameter quantization. More specifically, BLIP first trains a neural network with weight quantization on the new incoming task and then estimates information gain on each parameter provided by the task data to determine the bits to be frozen to prevent forgetting. We conduct extensive experiments ranging from classification tasks to reinforcement learning tasks, and the results show that our method produces better or on par results comparing to previous state-of-the-arts. Indeed, BLIP achieves close to zero forgetting while only requiring constant memory overheads throughout continual learning1.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Meta & Continual Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Meta & Continual Learning

Citations in the corpus

Online Continual Learning with Contrastive Vision Transformer (2022, methodology): “Regularizationbased methods limit the changes in important parameters during the learning of new tasks by estimating the importance of each network parameter for prior tasks [12,42,2,64,43,66,93].”
Learning to Imagine: Diversify Memory for Incremental Learning using Unlabeled Data (2022, methodology): “Popular structure-driven methods [2,20,21,25,32,33,35] modify the network structure or expand the network for training new tasks.”
Mimicking the Oracle: An Initial Phase Decorrelation Approach for Class Incremental Learning (2021, methodology): “Two classic setups of Incremental Learning are Class In-cremental Learning (CIL) [1,12,13,19,20,22,26] and Task Incremental Learning (TIL) [2, 16, 21, 24, 28, 30, 31].”
Recent Advances of Continual Learning in Computer Vision: An Overview (2021, positive): “[165] proposed a Bit-Level Information Preserving (BLIP) method, which updates the model parameters at the bit level by estimating information gain on each parameter.”
Continual Learning with Lifelong Vision Transformer (2022, methodology): “Regularization-based methods try to estimate the importance of each network parameter for prior tasks and penalize the changes of important parameters during the learning of new tasks [2,13,40,64,66,91].”
Continually Learning Self-Supervised Representations with Projected Functional Regularization (2021, background): “This can be done by either regularizing the weight space (constraining important parameters) [57, 60] or the functional space (constraining predictions or intermediate features) [13, 19, 31].”