Accelerated Linearized Laplace Approximation for Bayesian Deep Learning

Laplace approximation (LA) and its linearized variant (LLA) enable effortless adaptation of pretrained deep neural networks to Bayesian neural networks. The generalized Gauss-Newton (GGN) approximation is typically introduced to improve their tractability. However, LA and LLA are still confronted with non-trivial inefficiency issues and should rely on Kronecker-factored, diagonal, or even last-layer approximate GGN matrices in practical use. These approximations are likely to harm the fidelity of learning outcomes. To tackle this issue, inspired by the connections between LLA and neural target kernels (NTKs), we develop a Nystrom approximation to NTKs to accelerate LLA. Our method benefits from the capability of popular deep learning libraries for forward mode automatic differentiation, and enjoys reassuring theoretical guarantees. Extensive studies reflect the merits of the proposed method in aspects of both scalability and performance. Our method can even scale up to architectures like vision transformers. We also offer valuable ablation studies to diagnose our method. Code is available at https: //github. com/thudzj/ELLA.

Authors

• Zhijie Deng Shanghai Jiaotong University
• Feng Zhou Center for Applied Statistics and School of Statistics, Renmin University of China
• Jun Zhu Tsinghua University National Center for Protein Sciences (Beijing) Beijing Institute of Lifeomics

Keywords

No keywords are indexed for this paper.