Learning to Balance Local Losses via Meta-Learning

Seungdong Yoa, Minkyu Jeon, Youngjin Oh, Hyunwoo J. Kim

    Research output: Contribution to journalArticlepeer-review

    1 Citation (Scopus)

    Abstract

    The standard training for deep neural networks relies on a global and fixed loss function. For more effective training, dynamic loss functions have been recently proposed. However, the dynamic global loss function is not flexible to differentially train layers in complex deep neural networks. In this paper, we propose a general framework that learns to adaptively train each layer of deep neural networks via meta-learning. Our framework leverages the local error signals from layers and identifies which layer needs to be trained more at every iteration. Also, the proposed method improves the local loss function with our minibatch-wise dropout and cross-validation loop to alleviate meta-overfitting. The experiments show that our method achieved competitive performance compared to state-of-the-art methods on popular benchmark datasets for image classification: CIFAR-10 and CIFAR-100. Surprisingly, our method enables training deep neural networks without skip-connections using dynamically weighted local loss functions.

    Original languageEnglish
    Pages (from-to)130834-130844
    Number of pages11
    JournalIEEE Access
    Volume9
    DOIs
    Publication statusPublished - 2021

    Bibliographical note

    Funding Information:
    This work was supported in part by the Institute for Information and Communications Technology Planning and Evaluation (IITP) Grant by Korean Government through the Ministry of Science and ICT (MSIT) (Regional strategic industry convergence security core talent training business) under Grant 2019-0-01343, in part by the ICT Creative Consilience Program supervised by IITP under Grant IITP-2021-2020-0-01819, in part by the Electronics and Telecommunications Research Institute (ETRI) Grant by Korean Government (Fundamental technology research for human-centric autonomous intelligent systems) under Grant 21ZS1200, and in part by Korea Institute of Planning and Evaluation for Technology in Food, Agriculture and Forestry (IPET) and Korea Smart Farm Research and Development Foundation (KosFarm) through the Smart Farm Innovation Technology Development Program by the Ministry of Agriculture, Food and Rural Affairs (MAFRA), and MSIT, Rural Development Administration (RDA), under Grant 421025-04.

    Publisher Copyright:
    © 2013 IEEE.

    Keywords

    • Deep learning
    • image classification
    • machine learning
    • meta-learning

    ASJC Scopus subject areas

    • General Computer Science
    • General Materials Science
    • General Engineering

    Fingerprint

    Dive into the research topics of 'Learning to Balance Local Losses via Meta-Learning'. Together they form a unique fingerprint.

    Cite this