Ang, S.P. and Duong, S.T.M. and Phung, S.L. and Bouzerdoum, A. (2024) DAP: A dataset-agnostic predictor of neural network performance. Neurocomputing, 583. ISSN 09252312
Full text not available from this repository. (Upload)Abstract
Training a deep neural network on a large dataset to convergence is a time-demanding task. This task often must be repeated many times, especially when developing a new deep learning algorithm or performing a neural architecture search. This problem can be mitigated if a deep neural network's performance can be estimated without actually training it. In this work, we investigate the feasibility of two tasks: (i) predicting a deep neural network's performance accurately given only its architectural descriptor, and (ii) generalizing the predictor across different datasets without re-training. To this end, we propose a dataset-agnostic regression framework that uses a novel dual-LSTM model and a new dataset difficulty feature. The experimental results show that both tasks above are indeed feasible, and the proposed method outperforms the existing techniques in all experimental cases. Additionally, we also demonstrate several practical use-cases of the proposed predictor. © 2024 The Author(s)
Item Type: | Article |
---|---|
Divisions: | Offices > Office of International Cooperation |
Identification Number: | 10.1016/j.neucom.2024.127544 |
Uncontrolled Keywords: | Deep neural networks; Large datasets; Learning algorithms; Long short-term memory; Network architecture, Automl; Dataset-agnostic; Deep learning; Descriptors; Large datasets; Neural architecture search; Neural architectures; Neural network performance predictor; Neural-networks; Performance predictor, Network performance, Agnostic; article; deep learning; deep neural network; diagnosis; human; human experiment; learning algorithm; long short term memory network; nerve cell network |
URI: | http://eprints.lqdtu.edu.vn/id/eprint/11175 |