COLORA: Efficient Fine-Tuning for Convolutional Models with a Study Case on Optical Coherence Tomography Image Classification

2025/05/23 by Mariano Rivera, Rivera, Mariano, Angello Hoyos +1
Engineering · Medicine · #68T07 #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #I.1.2 #I.4.0 #I.4.10 #Optical Coherence Tomography Applications #Retinal Imaging and Analysis

paper · pdf · doi:10.48550/arxiv.2505.18315

openalex publication_date 2025/05/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We introduce CoLoRA (Convolutional Low-Rank Adaptation), a parameter-efficient fine-tuning method for convolutional neural networks (CNNs). CoLoRA extends LoRA to convolutional layers by decomposing kernel updates into lightweight depthwise and pointwise components.This design reduces the number of trainable parameters to 0.2 compared to conventional fine-tuning, preserves the original model size, and allows merging updates into the pretrained weights after each epoch, keeping inference complexity unchanged. On OCTMNISTv2, CoLoRA applied to VGG16 and ResNet50 achieves up to 1 percent accuracy and 0.013 AUC improvements over strong baselines (Vision Transformers, state-space, and Kolmogorov Arnold models) while reducing per-epoch training time by nearly 20 percent. Results indicate that CoLoRA provides a stable and effective alternative to full fine-tuning for medical image classification.

Citations

• MedKAN: An Advanced Kolmogorov-Arnold Network for Medical Image Classification
• Medical Image Classification with KAN-Integrated Transformers and Dilated Neighborhood Attention
• Adapters Strike Back
• Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey
• MedMamba: Vision Mamba for Medical Image Classification
• A Comprehensive Overview of Large Language Models
• Parameter-Efficient Fine-Tuning for Medical Image Analysis: The Missed Opportunity
• Medical SAM Adapter: Adapting Segment Anything Model for Medical Image Segmentation
• Micrograph segmentations for DDEVD
• Segment Anything
• A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT
• CHAPTER: Exploiting Convolutional Neural Network Adapters for Self-supervised Speech Models
• Towards a Unified View of Parameter-Efficient Transfer Learning
• Parameter-Efficient Transfer Learning for NLP
• Do Better ImageNet Models Transfer Better?
• Efficient parametrization of multi-domain deep neural networks
• Learning multiple visual domains with residual adapters
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Xception: Deep Learning with Depthwise Separable Convolutions
• Identity Mappings in Deep Residual Networks
• Deep Residual Learning for Image Recognition
• Going Deeper with Convolutions
• Very Deep Convolutional Networks for Large-Scale Image Recognition

Related

• Training Efficient CNNS: Tweaking the Nuts and Bolts of Neural Networks for Lighter, Faster and Robust Models
• AdaptoVision: A Multi-Resolution Image Recognition Model for Robust and Scalable Classification
• Best Practices for Convolutional Neural Networks Applied to Object Recognition in Images
• Speeding-up Convolutional Neural Networks Using Fine-tuned\n CP-Decomposition
• Convolutional Neural Networks: