TY  - EJOUR
AU  - Paul, Richard D.
AU  - Quercia, Alessio
AU  - Fortuin, Vincent
AU  - Nöh, Katharina
AU  - Scharr, Hanno
TI  - Parameter-efficient Bayesian Neural Networks for Uncertainty-aware Depth Estimation
PB  - arXiv
M1  - FZJ-2024-06731
PY  - 2024
N1  - Presented as an Extended Abstract at the 3rd Workshop on Uncertainty Quantification for Computer Vision at the ECCV'24.
AB  - State-of-the-art computer vision tasks, like monocular depth estimation (MDE), rely heavily on large, modern Transformer-based architectures. However, their application in safety-critical domains demands reliable predictive performance and uncertainty quantification. While Bayesian neural networks provide a conceptually simple approach to serve those requirements, they suffer from the high dimensionality of the parameter space. Parameter-efficient fine-tuning (PEFT) methods, in particular low-rank adaptations (LoRA), have emerged as a popular strategy for adapting large-scale models to down-stream tasks by performing parameter inference on lower-dimensional subspaces. In this work, we investigate the suitability of PEFT methods for subspace Bayesian inference in large-scale Transformer-based vision models. We show that, indeed, combining BitFit, DiffFit, LoRA, and CoLoRA, a novel LoRA-inspired PEFT method, with Bayesian inference enables more robust and reliable predictive performance in MDE.
KW  - Computer Vision and Pattern Recognition (cs.CV) (Other)
KW  - Machine Learning (stat.ML) (Other)
KW  - FOS: Computer and information sciences (Other)
LB  - PUB:(DE-HGF)25
DO  - DOI:10.48550/ARXIV.2409.17085
UR  - https://juser.fz-juelich.de/record/1033893
ER  -