RAC: a reusable adaptive convolution for CNN layer

Nguyen Viet Hung (East Asia University of Technology)
Phi Dinh Huynh (East Asia University of Technology)
Pham Hong Thinh (Quy Nhon University)
Phuc Hau Nguyen (Electric Power University)
Trong-Minh Hoang (Posts and Telecommunications Institute of Technology)

Indonesian Journal of Electrical Engineering and Computer Science
Vol 41, No 2: February 2026

This paper proposes reusable adaptive convolution (RAC), an efficient alternative to standard 3×3 convolutions for convolutional neural networks (CNNs). The main advantage of RAC lies in its simplicity and parameter efficiency, achieved by sharing horizontal and vertical 1×k/k×1 filter banks across blocks within a stage and recombining them through a lightweight 1×1 mixing layer. By operating at the operator design level, RAC avoids post-training compression steps and preserves the conventional Conv–BN–activation structure, enabling seamless integration into existing CNN backbones. To evaluate the effectiveness of the proposed method, extensive experiments are conducted on CIFAR-10 using several architectures, including ResNet-18/50/101, DenseNet, WideResNet, and EfficientNet. Experimental results demonstrate that RAC significantly reduces parameters and memory usage while maintaining competitive accuracy. These results indicate that RAC offers a reasonable balance between accuracy and compression, and is suitable for deploying CNN networks on resource-constrained platforms.