Meta-learning with a geometry-adaptive preconditioner

S Kang, D Hwang, M Eo, T Kim… - Proceedings of the …, 2023 - openaccess.thecvf.com
Abstract Model-agnostic meta-learning (MAML) is one of the most successful meta-learning
algorithms. It has a bi-level optimization structure where the outer-loop process learns a …

Few-shot classification via efficient meta-learning with hybrid optimization

J Jia, X Feng, H Yu - Engineering Applications of Artificial Intelligence, 2024 - Elsevier
Meta-learning is one of the important methods to solve the challenging few-shot learning
setting by using previous knowledge and experience to guide the learning of new tasks …

Multimodal parameter-efficient few-shot class incremental learning

M D'Alessandro, A Alonso… - Proceedings of the …, 2023 - openaccess.thecvf.com
Abstract Few-Shot Class Incremental Learning (FSCIL) is a challenging continual learning
task, where limited training examples are available during several learning sessions. To …

Generalizing supervised deep learning mri reconstruction to multiple and unseen contrasts using meta-learning hypernetworks

S Ramanarayanan, A Palla, K Ram… - Applied Soft …, 2023 - Elsevier
Meta-learning has recently been an emerging data-efficient learning technique for various
medical imaging operations and has helped advance contemporary deep learning models …

Integrating curriculum learning with meta-learning for general rhetoric identification

D Wang, Y Li, S Wang, X Li, X Chen, S Li… - International Journal of …, 2024 - Springer
Rhetoric is abundant and universal across different human languages. In this paper, we
propose a novel curriculum learning integrated with meta-learning (CLML) model to address …

Fast unsupervised deep outlier model selection with hypernetworks

X Ding, Y Zhao, L Akoglu - Proceedings of the 30th ACM SIGKDD …, 2024 - dl.acm.org
Deep neural network based Outlier Detection (DOD) has seen a recent surge of attention
thanks to the many advances in deep learning. In this paper, we consider a critical-yet …

The general framework for few-shot learning by kernel HyperNetworks

M Sendera, M Przewiȩźlikowski, J Miksa… - Machine Vision and …, 2023 - Springer
Few-shot models aim at making predictions using a minimal number of labeled examples
from a given task. The main challenge in this area is the one-shot setting, where only one …

Perspectives of Calibrated Adaptation for Few-shot Cross-domain Classification

D Kong, X Yang, N Wang, X Gao - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Current few-shot learning techniques predominantly leverage amortization techniques
based on meta-learning frameworks, which effectively adapt to unknown tasks with limited …

Hierarchical Adaptation with Hypernetworks for Few-shot Molecular Property Prediction

S Wu, Y Wang, Q Yao - arXiv preprint arXiv:2310.00614, 2023 - arxiv.org
Molecular property prediction (MPP) is important in biomedical applications, which naturally
suffers from a lack of labels, thus forming a few-shot learning problem. State-of-the-art …

DisRot: boosting the generalization capability of few-shot learning via knowledge distillation and self-supervised learning

C Ma, J Jia, J Huang, L Wu, X Wang - Machine Vision and Applications, 2024 - Springer
Few-shot learning (FSL) aims to adapt quickly to new categories with limited samples.
Despite significant progress in utilizing meta-learning for solving FSL tasks, challenges such …