Continual Learning with Low Rank Adaptation
This work addresses the challenge of retaining performance on old data while adapting to new data in continual learning, which is crucial for deploying AI systems in dynamic environments, though it is incremental as it builds on existing LoRA techniques.
The paper tackles the problem of catastrophic forgetting in continual learning for pre-trained transformers by applying Low Rank Adaptation (LoRA), achieving state-of-the-art performance on domain-incremental benchmarks while maintaining parameter efficiency comparable to prompt tuning methods.
Recent work using pretrained transformers has shown impressive performance when fine-tuned with data from the downstream problem of interest. However, they struggle to retain that performance when the data characteristics changes. In this paper, we focus on continual learning, where a pre-trained transformer is updated to perform well on new data, while retaining its performance on data it was previously trained on. Earlier works have tackled this primarily through methods inspired from prompt tuning. We question this choice, and investigate the applicability of Low Rank Adaptation (LoRA) to continual learning. On a range of domain-incremental learning benchmarks, our LoRA-based solution, CoLoR, yields state-of-the-art performance, while still being as parameter efficient as the prompt tuning based methods.