ATTL: An Automated Targeted Transfer Learning with Deep Neural Networks
Authors:
- 1 December 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Abstract
Success of machine learning algorithms hinges on access to labeled dataset. Obtaining a labeled dataset is an expensive, challenging and time-consuming process, leading to the development of transfer learning (TL) methodology. TL incorporates gained knowledge from a previously trained source model into specific yet similar task models with limited data domain coverage. In this paper, we propose an automated targeted transfer learning (ATTL) method to resolve the transferability between source and target with minimal data requirements. The ATTL method decides how much target data is essential for model training, along with selected source data, to obtain the skateholder's specified performance metrics. The ATTL framework optimizes the system to select minimal target data based on two approaches: combinatorial coverage and adaptive selection methodology, along with specific source data for fine-tuning given a pre-trained source model. We evaluated the ATTL method on the Kaggle's ‘planes in satellite imagery’ dataset and the results identified that acquiring a small number of intentionally well-chosen samples from the target environment can achieve model performance of 97% in comparison to the baseline transfer learning accuracy of 92%.Keywords
Funding Information
- U.S. Army Research Laboratory
This publication has 11 references indexed in Scilit:
- AdaFilter: Adaptive Filter Fine-Tuning for Deep Transfer LearningProceedings of the AAAI Conference on Artificial Intelligence, 2020
- A Flexible Selection Scheme for Minimum-Effort Transfer Learning2020
- A Deep Learning Framework for Automated Transfer Learning of Neural Networks2019
- A Systems Theoretic Perspective on Transfer Learning2019
- Smart finite elements: A novel machine learning applicationComputer Methods in Applied Mechanics and Engineering, 2019
- Transfer learning for detecting unknown network attacksEURASIP Journal on Information Security, 2019
- Convolutional neural networks: an overview and application in radiologyInsights into Imaging, 2018
- On automated source selection for transfer learning in convolutional neural networksPattern Recognition, 2018
- Borrowing Treasures from the Wealthy: Deep Transfer Learning through Selective Joint Fine-Tuning2017
- Combinatorial Coverage Measurement Concepts and Applications2013