Home | Publications | CDS+24

Variational Low-Rank Adaptation Using IVON

MCML Authors

Yuesong Shen

Dr.

* Former Member

→ Group Daniel Cremers
Computer Vision & Artificial Intelligence

Daniel Cremers

Prof. Dr.

Director

Computer Vision & Artificial Intelligence

Abstract

We show that variational learning can significantly improve the accuracy and calibration of Low-Rank Adaptation (LoRA) without a substantial increase in the cost. We replace AdamW by the Improved Variational Online Newton (IVON) algorithm to finetune large language models. For Llama-2 with 7 billion parameters, IVON improves the accuracy over AdamW by 2.8% and expected calibration error by 4.6%. The accuracy is also better than the other Bayesian alternatives, yet the cost is lower and the implementation is easier. Our work provides additional evidence for the effectiveness of IVON for large language models.

inproceedings

FITML @NeurIPS 2024

Workshop Fine-Tuning in Modern Machine Learning: Principles and Scalability at the 38th Conference on Neural Information Processing Systems. Vancouver, Canada, Dec 10-15, 2024.