Classifier Transfer

Classifier transfer focuses on adapting pre-trained machine learning models to new tasks or domains, aiming to improve efficiency and performance compared to training from scratch. Current research emphasizes efficient model selection strategies, exploring techniques to identify the most suitable pre-trained model for a given task and investigating the impact of model architecture on transferability, including the use of Support Vector Machines and deep neural networks. These advancements are significant for various applications, such as improving the performance of deep learning on limited datasets (e.g., SAR imagery) and enabling more robust and adaptable classifiers in domains with data shifts or class imbalances (e.g., brain-computer interfaces).

Papers

March 11, 2024

Pre-Trained Model Recommendation for Downstream Fine-tuning
Jiameng Bai, Sai Wu, Jie Song, Junbo Zhao, Gang Chen
Transfer Learning Pre Trained Model Recommender System Large Vision Model Vector Valued Downstream Fine Tuning Classifier Transfer

February 28, 2024

Deep Neural Network Models Trained With A Fixed Random Classifier Transfer Better Across Domains
Hafiz Tiomoko Ali, Umberto Michieli, Ji Joong Moon, Daehyun Kim, Mete Ozay
Deep Neural Network DNN Model Neural Collapse Domain Name Covariance Regularization Classifier Transfer

October 26, 2023

Deep Learning on SAR Imagery: Transfer Learning Versus Randomly Initialized Weights
Morteza Karimzadeh, Rafael Pires de Lima
Deep Learning Deep Learning Model Pre Trained Model Synthetic Aperture Radar Balancing Weight Classifier Transfer

August 10, 2022

Classifier Transfer with Data Selection Strategies for Online Support Vector Machine Classification with Class Imbalance
Mario Michael Krell, Nils Wilshusen, Anett Seeland, Su Kyoung Kim
Class Imbalance Support Vector Machine Data Selection Dataset Shift Online Classification Classifier Transfer

Classifier Transfer

Papers

Pre-Trained Model Recommendation for Downstream Fine-tuning

Deep Neural Network Models Trained With A Fixed Random Classifier Transfer Better Across Domains

Deep Learning on SAR Imagery: Transfer Learning Versus Randomly Initialized Weights

Classifier Transfer with Data Selection Strategies for Online Support Vector Machine Classification with Class Imbalance