paper-with-me

홈 › Papers

How stable are Transferability Metrics evaluations?

2022-04-04 · Andrea Agostinelli, Michal Pándy, Jasper Uijlings, Thomas Mensink, Vittorio Ferrari

Transferability metrics is a maturing field with increasing interest, which aims at providing heuristics for selecting the most suitable source models to transfer to a given target dataset, without fine-tuning them all. However, existing works rely on custom experimental setups which differ across papers, leading to inconsistent conclusions about which transferability metrics work best. In this paper we conduct a large-scale study by systematically constructing a broad range of 715k experimental setup variations. We discover that even small variations to an experimental setup lead to different conclusions about the superiority of a transferability metric over another. Then we propose better evaluations by aggregating across many experiments, enabling to reach more stable conclusions. As a result, we reveal the superiority of LogME at selecting good source datasets to transfer from in a semantic segmentation scenario, NLEEP at selecting good source architectures in an image classification scenario, and GBC at determining which target task benefits most from a given source model. Yet, no single transferability metric works best in all scenarios.

📄 PDF Abstract BibTeX arXiv:2204.01403

Code (0)

등록된 구현이 없습니다.

Tasks

image-classificationImage ClassificationSemantic Segmentation

Similar Papers 제목 키워드 기반

Improving Adversarial Transferability by Stable Diffusion

2023-11-18 · Jiayang Liu, Siyu Zhu, Siyuan Liang, Jie Zhang 외

Deep neural networks (DNNs) are susceptible to adversarial examples, which introduce imperceptible perturbations to benign samples, deceiving DNN predictions. While some attack methods excel in the white-box setting, the…

Towards Good Practices in Evaluating Transfer Adversarial Attacks

2022-11-17 · Zhengyu Zhao, Hanwei Zhang, Renjue Li, Ronan Sicre 외

Transfer adversarial attacks raise critical security concerns in real-world, black-box scenarios. However, the actual progress of this field is difficult to assess due to two common limitations in existing evaluations. F…

Transferability Metrics for Object Detection

2023-06-27 · Louis Fouquet, Simona Maggio, Léo Dreyfus-Schmidt

Transfer learning aims to make the most of existing pre-trained models to achieve better performance on a new task in limited data scenarios. However, it is unclear which models will perform best on which task, and it is…

Objectobject-detectionObject DetectionTransfer Learning

Occam's model: Selecting simpler representations for better transferability estimation

2025-02-10 · Prabhant Singh, Sibylle Hess, Joaquin Vanschoren

Fine-tuning models that have been pre-trained on large datasets has become a cornerstone of modern machine learning workflows. With the widespread availability of online model repositories, such as Hugging Face, it is no…

Understanding Knowledge Transferability for Transfer Learning: A Survey

2025-07-03 · Haohua Wang, Jingge Wang, Zijie Zhao, Yang Tan 외 arxiv

Transfer learning has become an essential paradigm in artificial intelligence, enabling the transfer of knowledge from a source task to improve performance on a target task. This approach, particularly through techniques…

Transfer Learning