Sequential Function-Space Variational Inference via Gaussian Mixture Approximation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Menghao Waiyan William, Hao, Pengcheng, Kuruoğlu, Ercan Engin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916761222250496
author Zhu, Menghao Waiyan William
Hao, Pengcheng
Kuruoğlu, Ercan Engin
author_facet Zhu, Menghao Waiyan William
Hao, Pengcheng
Kuruoğlu, Ercan Engin
contents Continual learning in neural networks aims to learn new tasks without forgetting old tasks. Sequential function-space variational inference (SFSVI) uses a Gaussian variational distribution to approximate the distribution of the outputs of the neural network corresponding to a finite number of selected inducing points. Since the posterior distribution of a neural network is multi-modal, a Gaussian distribution could only match one mode of the posterior distribution, and a Gaussian mixture distribution could be used to better approximate the posterior distribution. We propose an SFSVI method based on a Gaussian mixture variational distribution. We also compare different types of variational inference methods with a fixed pre-trained feature extractor (where continual learning is performed on the final layer) and without a fixed pre-trained feature extractor (where continual learning is performed on all layers). We find that in terms of final average accuracy, likelihood-focused Gaussian mixture SFSVI outperforms other sequential variational inference methods, especially in the latter case.
format Preprint
id arxiv_https___arxiv_org_abs_2503_07114
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Sequential Function-Space Variational Inference via Gaussian Mixture Approximation
Zhu, Menghao Waiyan William
Hao, Pengcheng
Kuruoğlu, Ercan Engin
Machine Learning
Continual learning in neural networks aims to learn new tasks without forgetting old tasks. Sequential function-space variational inference (SFSVI) uses a Gaussian variational distribution to approximate the distribution of the outputs of the neural network corresponding to a finite number of selected inducing points. Since the posterior distribution of a neural network is multi-modal, a Gaussian distribution could only match one mode of the posterior distribution, and a Gaussian mixture distribution could be used to better approximate the posterior distribution. We propose an SFSVI method based on a Gaussian mixture variational distribution. We also compare different types of variational inference methods with a fixed pre-trained feature extractor (where continual learning is performed on the final layer) and without a fixed pre-trained feature extractor (where continual learning is performed on all layers). We find that in terms of final average accuracy, likelihood-focused Gaussian mixture SFSVI outperforms other sequential variational inference methods, especially in the latter case.
title Sequential Function-Space Variational Inference via Gaussian Mixture Approximation
topic Machine Learning
url https://arxiv.org/abs/2503.07114