Close menu

SURE

Sunderland Repository records the research produced by the University of Sunderland including practice-based research and theses.

Enhancing Transfer Learning Reliability via Block-wise Fine-tuning

Barakat, Basel and Huang, Qiang (2023) Enhancing Transfer Learning Reliability via Block-wise Fine-tuning. In: 22nd IEEE International Conference on Machine Learning and Applications ICMLA 2023. IEEE, pp. 414-421. ISBN 979-8-3503-4534-6

Item Type: Book Section

Abstract

Fine-tuning can be used to tackle domain specific tasks by transferring knowledge learned from
pre-trained models.
However, previous studies on fine-tuning focused on adapting only the weights of a task-specific classifier or re-optimising all layers of the pre-trained model using the new task data.
The first type of method cannot mitigate the mismatch between a pre-trained model and the
new task data, and the second type of method easily causes over-fitting when processing tasks with limited data.
To explore the effectiveness of fine-tuning, we propose a novel block-wise optimisation mechanism, which adapts the weights of a group of layers of a pre-trained model.
This work presents a theoretical framework and empirical evaluation of block-wise fine-tuning to find a reliable transfer learning strategy.
The proposed approach is evaluated on two datasets, Oxford Flowers and Caltech 101, using 15 commonly used state-of-the-art pre-trained base models.

Results indicate that the proposed strategy consistently outperforms the baselines in terms of classification accuracy, although the specific block leading to optimal performance may vary across models.
The investigation reveals that selecting a block from the fourth quarter of a base model generally yields improved performance compared to the baselines.
Overall, the block-wise approach consistently outperforms the baselines and exhibits higher accuracy and reliability.
This study provides valuable insights into the selection of salient blocks and highlights the effectiveness of block-wise fine-tuning in achieving improved classification accuracy in various models and datasets.

[img] PDF
Enhancing_Transfer_Learning_Reliability_via_Block_wise_Fine_tuning.pdf - Accepted Version
Restricted to Repository staff only

Download (1MB) | Request a copy

More Information

Depositing User: Basel Barakat

Identifiers

Item ID: 16762
Identification Number: https://doi.org/10.1109/ICMLA58977.2023
ISBN: 979-8-3503-4534-6
URI: http://sure.sunderland.ac.uk/id/eprint/16762
Official URL: https://ieeexplore.ieee.org/xpl/conhome/10459339/p...

Users with ORCIDS

ORCID for Basel Barakat: ORCID iD orcid.org/0000-0001-9126-7613
ORCID for Qiang Huang: ORCID iD orcid.org/0000-0002-2943-2283

Catalogue record

Date Deposited: 18 Dec 2023 10:57
Last Modified: 01 Oct 2024 09:00

Contributors

Author: Basel Barakat ORCID iD
Author: Qiang Huang ORCID iD

University Divisions

Faculty of Technology > School of Computer Science

Subjects

Computing > Artificial Intelligence

Actions (login required)

View Item (Repository Staff Only) View Item (Repository Staff Only)