Continual Learning through Retrieval and Imagination

By Zhen Wang, Liu Liu, Yiqun Duan, Dacheng Tao. 2022

48 citations recorded by the source corpus.

Continual learning is an intellectual ability of artificial agents to learn new streaming labels from sequential data. The main impediment to continual learning is catastrophic forgetting, a severe performance degradation on previously learned tasks. Although simply replaying all previous data or continuously adding the model parameters could alleviate the issue, it is impractical in real-world applications due to the limited available resources. Inspired by the mechanism of the human brain to deepen its past impression, we propose a novel framework, Deep Retrieval and Imagination (DRI), which consists of two components: 1) an embedding network that constructs a unified embedding space without adding model parameters on the arrival of new tasks; and 2) a generative model to produce additional (imaginary) data based on the limited memory. By retrieving the past experiences and corresponding imaginary data, DRI distills knowledge and rebalances the embedding space to further mitigate forgetting. Theoretical analysis demonstrates that DRI can reduce the loss approximation error and improve the robustness through retrieval and imagination, bringing better generalizability to the network. Extensive experiments show that DRI performs significantly better than the existing state-of-the-art continual learning methods and effectively alleviates catastrophic forgetting.

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

1. Hang Su
2. Jun Zhu
3. Liyuan Wang
4. Xingxing Zhang
5. Liu Liu
6. Prashant Bhat
7. Bahram Zonooz
8. Elahe Arani

In Computer Vision and Pattern Recognition

1. Hang Su
2. Jun Zhu
3. Liyuan Wang
4. Xingxing Zhang
5. Dacheng Tao
6. Liu Liu
7. Zhen Wang
8. Yajing Kong

In Meta & Continual Learning

1. Xingxing Zhang
2. Liyuan Wang
3. Hang Su
4. Jun Zhu
5. Prashant Bhat
6. Bahram Zonooz
7. Elahe Arani
8. Yajing Kong

Citations in the corpus

• A Comprehensive Survey of Continual Learning: Theory, Method and Application (2023, background): “…tasksolutions[52],[104],[113],[154],[196],balancedexploita-tionofoldandnewtrainingsamples[46],[129],[147],[148], [149],[152],spacereservationforsubsequenttasks[115],[116], [256],etc. Ontheotherhand,solutiongeneralizabilitycould…”
• Task-Aware Information Routing from Common Representation Space in Lifelong Learning (2023, methodology): “In addition to rehearsal, DRI (Wang et al., 2022) utilizes a generative model to augment rehearsal under low buffer regimes.”
• Balancing Stability and Plasticity through Advanced Null Space in Continual Learning (2022, neutral): “Replayed-based methods leverage episodic memory to store representative history data or generate virtual data via a generative model, and replay these samples with current data [2, 7, 8, 12, 26, 37, 39, 40, 45, 50, 51].”
• Online Continual Learning with Contrastive Vision Transformer (2022, background): “Rehearsal-based methods [8,37,81,60,15,22,49,79,68,52] alleviate catastrophic forgetting by replaying a subset of data of past tasks stored in limited buffer.”