Bayesian Parameter-Efficient Fine-Tuning for Overcoming Catastrophic Forgetting

arXiv:2402.12220v1 Announce Type: cross
Abstract: Although motivated by the adaptation of text-to-speech synthesis models, we argue that more generic parameter-efficient fine-tuning (PEFT) is an appropriate framework to do such adaptation. However, 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 …

abstract arxiv bayesian capabilities catastrophic forgetting cs.lg eess.as fine-tuning framework issue peft speech synthesis text text-to-speech type