From Deep Additive Kernel Learning to Last-Layer Bayesian Neural Networks via Induced Prior Approximation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhao, Wenyuan, Chen, Haoyuan, Liu, Tie, Tuo, Rui, Tian, Chao
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915152759095296
author Zhao, Wenyuan
Chen, Haoyuan
Liu, Tie
Tuo, Rui
Tian, Chao
author_facet Zhao, Wenyuan
Chen, Haoyuan
Liu, Tie
Tuo, Rui
Tian, Chao
contents With the strengths of both deep learning and kernel methods like Gaussian Processes (GPs), Deep Kernel Learning (DKL) has gained considerable attention in recent years. From the computational perspective, however, DKL becomes challenging when the input dimension of the GP layer is high. To address this challenge, we propose the Deep Additive Kernel (DAK) model, which incorporates i) an additive structure for the last-layer GP; and ii) induced prior approximation for each GP unit. This naturally leads to a last-layer Bayesian neural network (BNN) architecture. The proposed method enjoys the interpretability of DKL as well as the computational advantages of BNN. Empirical results show that the proposed approach outperforms state-of-the-art DKL methods in both regression and classification tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2502_10540
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle From Deep Additive Kernel Learning to Last-Layer Bayesian Neural Networks via Induced Prior Approximation
Zhao, Wenyuan
Chen, Haoyuan
Liu, Tie
Tuo, Rui
Tian, Chao
Machine Learning
With the strengths of both deep learning and kernel methods like Gaussian Processes (GPs), Deep Kernel Learning (DKL) has gained considerable attention in recent years. From the computational perspective, however, DKL becomes challenging when the input dimension of the GP layer is high. To address this challenge, we propose the Deep Additive Kernel (DAK) model, which incorporates i) an additive structure for the last-layer GP; and ii) induced prior approximation for each GP unit. This naturally leads to a last-layer Bayesian neural network (BNN) architecture. The proposed method enjoys the interpretability of DKL as well as the computational advantages of BNN. Empirical results show that the proposed approach outperforms state-of-the-art DKL methods in both regression and classification tasks.
title From Deep Additive Kernel Learning to Last-Layer Bayesian Neural Networks via Induced Prior Approximation
topic Machine Learning
url https://arxiv.org/abs/2502.10540