Saved in:
Bibliographic Details
Main Authors: Dai, Gaole, Xu, Huatao, Yang, Yifan, Tan, Rui, Li, Mo
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.04060
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913682523422720
author Dai, Gaole
Xu, Huatao
Yang, Yifan
Tan, Rui
Li, Mo
author_facet Dai, Gaole
Xu, Huatao
Yang, Yifan
Tan, Rui
Li, Mo
contents Expanding existing learning systems to provide high-quality customized models for more domains, such as new users, is challenged by the limited labeled data and the data and device heterogeneities. While knowledge distillation methods could overcome label scarcity and device heterogeneity, they assume the teachers are fully reliable and overlook the data heterogeneity, which prevents the direct adoption of existing models. To address this problem, this paper proposes a framework, HaT, to expand learning systems. It first selects multiple high-quality models from the system at a low cost and then fuses their knowledge by assigning sample-wise weights to their predictions. Later, the fused knowledge is selectively injected into the customized models based on the knowledge quality. Extensive experiments on different tasks, modalities, and settings show that HaT outperforms state-of-the-art baselines by up to 16.5% accuracy and saves up to 39% communication traffic.
format Preprint
id arxiv_https___arxiv_org_abs_2412_04060
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Expand Heterogeneous Learning Systems with Selective Multi-Source Knowledge Fusion
Dai, Gaole
Xu, Huatao
Yang, Yifan
Tan, Rui
Li, Mo
Artificial Intelligence
Expanding existing learning systems to provide high-quality customized models for more domains, such as new users, is challenged by the limited labeled data and the data and device heterogeneities. While knowledge distillation methods could overcome label scarcity and device heterogeneity, they assume the teachers are fully reliable and overlook the data heterogeneity, which prevents the direct adoption of existing models. To address this problem, this paper proposes a framework, HaT, to expand learning systems. It first selects multiple high-quality models from the system at a low cost and then fuses their knowledge by assigning sample-wise weights to their predictions. Later, the fused knowledge is selectively injected into the customized models based on the knowledge quality. Extensive experiments on different tasks, modalities, and settings show that HaT outperforms state-of-the-art baselines by up to 16.5% accuracy and saves up to 39% communication traffic.
title Expand Heterogeneous Learning Systems with Selective Multi-Source Knowledge Fusion
topic Artificial Intelligence
url https://arxiv.org/abs/2412.04060