paper-with-me

홈 › Papers

PACTran: PAC-Bayesian Metrics for Estimating the Transferability of Pretrained Models to Classification Tasks

2022-03-10 · Nan Ding, Xi Chen, Tomer Levinboim, Beer Changpinyo, Radu Soricut

With the increasing abundance of pretrained models in recent years, the problem of selecting the best pretrained checkpoint for a particular downstream classification task has been gaining increased attention. Although several methods have recently been proposed to tackle the selection problem (e.g. LEEP, H-score), these methods resort to applying heuristics that are not well motivated by learning theory. In this paper we present PACTran, a theoretically grounded family of metrics for pretrained model selection and transferability measurement. We first show how to derive PACTran metrics from the optimal PAC-Bayesian bound under the transfer learning setting. We then empirically evaluate three metric instantiations of PACTran on a number of vision tasks (VTAB) as well as a language-and-vision (OKVQA) task. An analysis of the results shows PACTran is a more consistent and effective transferability measure compared to existing selection methods.

📄 PDF Abstract BibTeX arXiv:2203.05126

Code (1)

google-research/pactran_metrics 공식 구현 tf

Tasks

Learning TheoryModel SelectionTransferabilityTransfer Learning

Similar Papers 제목 키워드 기반

Building a Winning Team: Selecting Source Model Ensembles using a Submodular Transferability Estimation Approach

2023-09-05 · ICCV 2023 1 · Vimal K B, Saketh Bachu, Tanmay Garg, Niveditha Lakshmi Narasimhan 외

Estimating the transferability of publicly available pretrained models to a target task has assumed an important place for transfer learning tasks in recent years. Existing efforts propose metrics that allow a user to ch…

image-classificationImage ClassificationSemantic SegmentationTransfer 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…

Towards Estimating Transferability using Hard Subsets

2023-01-17 · Tarun Ram Menta, Surgan Jandial, Akash Patil, Vimal KB 외

As transfer learning techniques are increasingly used to transfer knowledge from the source model to the target task, it becomes important to quantify which source models are suitable for a given target task without perf…

Transfer Learning

How to Estimate Model Transferability of Pre-Trained Speech Models?

2023-06-01 · Zih-Ching Chen, Chao-Han Huck Yang, Bo Li, Yu Zhang 외

In this work, we introduce a "score-based assessment" framework for estimating the transferability of pre-trained speech models (PSMs) for fine-tuning target tasks. We leverage upon two representation theories, Bayesian …

On dataset transferability in medical image classification

2024-12-28 · Dovile Juodelyte, Enzo Ferrante, Yucheng Lu, Prabhant Singh 외

Current transferability estimation methods designed for natural image datasets are often suboptimal in medical image classification. These methods primarily focus on estimating the suitability of pre-trained source model…

BenchmarkingClassificationimage-classificationImage Classification+1