AI Navigate

分割学習における適応チャネルプルーニングのためのラベル認識チャネルスコアリングの活用

arXiv cs.LG / 2026/3/11

Ideas & Deep AnalysisModels & Research

要点

  • 本論文は、クライアントとサーバ間で中間特徴表現(スマッシュデータ)を送信することで発生する分割学習(SL)の通信オーバーヘッドに対処する。
  • ラベル認識チャネル重要度スコアリング(LCIS)モジュールによって重要度の低いチャネルを識別し、適応チャネルプルーニング(ACP)スキームを提案する。
  • このプルーニングにより送信されるスマッシュデータのサイズを削減し、モデル性能を損なうことなく通信オーバーヘッドを効果的に低減する。
  • 実験結果は、ACP-SLがベンチマーク手法に比べテスト精度を向上させ、より速く収束することで通信負荷をさらに軽減できることを示している。
  • この手法は分散学習における計算負荷と通信効率のバランスを実現し、多数のクライアントデバイスが存在するシナリオでのスケーラビリティを向上させる。

Computer Science > Machine Learning

arXiv:2603.09792 (cs)
[Submitted on 10 Mar 2026]

Title:Exploiting Label-Aware Channel Scoring for Adaptive Channel Pruning in Split Learning

View a PDF of the paper titled Exploiting Label-Aware Channel Scoring for Adaptive Channel Pruning in Split Learning, by Jialei Tan and 6 other authors
View PDF HTML (experimental)
Abstract:Split learning (SL) transfers most of the training workload to the server, which alleviates computational burden on client devices. However, the transmission of intermediate feature representations, referred to as smashed data, incurs significant communication overhead, particularly when a large number of client devices are involved. To address this challenge, we propose an adaptive channel pruning-aided SL (ACP-SL) scheme. In ACP-SL, a label-aware channel importance scoring (LCIS) module is designed to generate channel importance scores, distinguishing important channels from less important ones. Based on these scores, an adaptive channel pruning (ACP) module is developed to prune less important channels, thereby compressing the corresponding smashed data and reducing the communication overhead. Experimental results show that ACP-SL consistently outperforms benchmark schemes in test accuracy. Furthermore, it reaches a target test accuracy in fewer training rounds, thereby reducing communication overhead.
Comments:
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Cite as: arXiv:2603.09792 [cs.LG]
  (or arXiv:2603.09792v1 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2603.09792
Focus to learn more
arXiv-issued DOI via DataCite

Submission history

From: Lin Zheng [view email]
[v1] Tue, 10 Mar 2026 15:25:08 UTC (967 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Exploiting Label-Aware Channel Scoring for Adaptive Channel Pruning in Split Learning, by Jialei Tan and 6 other authors
  • View PDF
  • HTML (experimental)
  • TeX Source
Current browse context:
cs.LG
< prev   |   next >
Change to browse by:

References & Citations

export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo
Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
Links to Code Toggle
Papers with Code (What is Papers with Code?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos

Demos

Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers

Recommenders and Search Tools

Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
IArxiv recommender toggle
IArxiv Recommender (What is IArxiv?)
About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.