Enhancing Model Agnostic Meta-Learning via Gradient Similarity Loss
-
Published:2024-01-29
Issue:3
Volume:13
Page:535
-
ISSN:2079-9292
-
Container-title:Electronics
-
language:en
-
Short-container-title:Electronics
Author:
Tak Jae-Ho1ORCID, Hong Byung-Woo1
Affiliation:
1. Department of Artificial Intelligence, Chung-Ang University, Seoul 156-756, Republic of Korea
Abstract
Artificial intelligence (AI) technology has advanced significantly, now capable of performing tasks previously believed to be exclusive to skilled humans. However, AI models, in contrast to humans who can develop skills with relatively less data, often require substantial amounts of data to emulate human cognitive abilities in specific areas. In situations where adequate pre-training data is not available, meta-learning becomes a crucial method for enhancing generalization. The Model Agnostic Meta-Learning (MAML) algorithm, which employs second-order derivative calculations to fine-tune initial parameters for better starting points, plays a pivotal role in this area. However, the computational demand of this method can be challenging for modern models with a large number of parameters. The concept of the Approximate Hessian Effect is introduced in this context, examining the effectiveness of second-order derivatives in identifying initial parameters conducive to high generalization performance. The study suggests the use of cosine similarity and squared error (L2 loss) as a loss function within the Approximate Hessian Effect framework to modify gradient weights, aiming for more generalizable model parameters. Additionally, an algorithm that relies on first-order calculations is presented, designed to achieve performance levels comparable to MAML. This approach was tested and compared with traditional MAML methods using both the MiniImagenet dataset and a modified MNIST dataset. The results were analyzed to evaluate its efficiency. Compared to previous studies that achieved good performance using only the first derivative, this approach is more efficient because it does not require iterative loops to converge on additional loss functions. Additionally, there is potential for further performance enhancement through hyperparameter tuning.
Funder
Chung-Ang University Research Scholarship Grants in 2022 Korea Government National Research Foundation of Korea
Subject
Electrical and Electronic Engineering,Computer Networks and Communications,Hardware and Architecture,Signal Processing,Control and Systems Engineering
Reference33 articles.
1. Matching networks for one-shot learning;Vinyals;Adv. Neural Inf. Process. Syst.,2016 2. Meta-Learning in Neural Networks: A Survey;Hospedales;IEEE Trans. Pattern Anal. Mach. Intell.,2022 3. A survey of deep meta-learning;Huisman;Artif. Intell. Rev.,2021 4. Achille, A., Lam, M., Tewari, R., Ravichandran, A., Maji, S., Fowlkes, C.C., Soatto, S., and Perona, P. (November, January 27). Task2Vec: Task Embedding for Meta-Learning. Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Republic of Korea. 5. Wu, Z., Shi, X., Lin, G., and Cai, J. (2021, January 11–17). Learning Meta-class Memory for Few-Shot Semantic Segmentation. Proceedings of the IEEE/CVF International Conference on Computer Vision, Montreal, BC, Canada.
|
|