Bayesian Parameter-Efficient Fine-Tuning for Overcoming Catastrophic Forgetting
| Type of publication: | Journal paper |
| Citation: | Chen_TASLP_2024 |
| Publication status: | Accepted |
| Journal: | IEEE/ACM Transactions on Audio, Speech, and Language Processing |
| Year: | 2024 |
| DOI: | 10.1109/TASLP.2024.3463395 |
| Abstract: | We are motivated primarily by the adaptation of text-to-speech synthesis models; however we argue that more generic parameter-efficient fine-tuning (PEFT) is an appropriate framework to do such adaptation. Nevertheless, catastrophic forgetting remains an issue with PEFT, damaging the pre-trained model's inherent capabilities. We demonstrate that existing Bayesian learning techniques can be applied to PEFT to prevent catastrophic forgetting as long as the parameter shift of the fine-tuned layers can be calculated differentiably. In a principled series of experiments on language modeling and speech synthesis tasks, we utilize established Laplace approximations, including diagonal and Kronecker-factored approaches, to regularize PEFT with the low-rank adaptation (LoRA) and compare their performance in pre-training knowledge preservation. Our results demonstrate that catastrophic forgetting can be overcome by our methods without degrading the fine-tuning performance, and using the Kronecker-factored approximation produces a better preservation of the pre-training knowledge than the diagonal ones. |
| Keywords: | Bayesian transfer learning, catastrophic forgetting, Laplace approximation, parameter-efficient fine-tuning |
| Projects |
Idiap NAST |
| Authors | |
| Added by: | [UNK] |
| Total mark: | 0 |
|
Attachments
|
|
|
|
|
Notes
|
|
|
|
|
