cs.DCSep 30, 2025

Hybrid Dual-Batch and Cyclic Progressive Learning for Efficient Distributed Training of Convolutional Neural Networks

Authors: Kuan-Wei Lu, Ding-Yong Hong, Pangfeng Liu, Jan-Jan Wu

Organizations: Institute of Information Science, Academia Sinica, Taipei, Taiwan · Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan

Abstract

Distributed deep learning is critical for training convolutional neural network (CNN) models on large-scale image datasets with numerous parameters. Current research primarily focuses on leveraging additional hardware resources and powerful computing units to accelerate the training process. As a result, larger batch sizes are often employed to speed up training. However, training with large batch sizes can lead to lower accuracy due to poor generalization. To address this issue, we propose the dual-batch learning scheme, a distributed training method built on the parameter server framework. This approach maximizes training efficiency by utilizing the largest batch size that the hardware can support while incorporating a smaller batch size to enhance model generalization. By using two different batch sizes simultaneously, this method improves accuracy with minimal additional training time. Additionally, to mitigate the time overhead caused by dual-batch learning, we propose the cyclic progressive learning scheme. This technique repeatedly and gradually increases image resolution from low to high during training, thereby reducing training time. By combining cyclic progressive learning with dual-batch learning, our hybrid approach improves both model generalization and training efficiency for image classification tasks. Experimental results with ResNet-18 demonstrate that, compared to conventional training methods, our approach improves accuracy by 3.3% while reducing training time by 10.1% on CIFAR-100, and further achieves a 34.7% reduction in training time on ImageNet.

Figures & tables

Explore similar work

CardsList
  1. Learn&Drop: Fast Learning of CNNs based on Layer Dropping

    Apr 25, 2026Giorgio Cruciata, Luca Cruciata, Liliana Lo Presti +2Deep Learning OptimizationConvolutional Neural Networks

  2. Learning Faster without Deeper Networks: A*-Inspired Batch Selection for Efficient CNN Training

    Jul 17, 2026Anxhelo Shehu, Enes Stastoli, Arben CelaDeep Learning OptimizationConvolutional Neural Networks

  3. Tula: Optimizing Time, Cost, and Generalization in Distributed Large-Batch Training

    Mar 18, 2026Sahil Tyagi, Feiyi WangDistributed TrainingHyperparameter Optimization