Metadiff: Meta-learning with conditional diffusion for few-shot learning

B Zhang, C Luo, D Yu, X Li, H Lin, Y Ye… - Proceedings of the AAAI …, 2024 - ojs.aaai.org
Equipping a deep model the ability of few-shot learning (FSL) is a core challenge for artificial
intelligence. Gradient-based meta-learning effectively addresses the challenge by learning …

Few-shot intent detection with self-supervised pretraining and prototype-aware attention

S Yang, YJ Du, X Zheng, XY Li, XL Chen, YL Li… - Pattern Recognition, 2024 - Elsevier
Few-shot intent detection is a more challenging application. However, traditional prototypical
networks based on averaging often suffer from issues such as missing key information, poor …

Few-shot multi-domain text intent classification with Dynamic Balance Domain Adaptation Meta-learning

S Yang, YJ Du, J Liu, XY Li, XL Chen, HM Gao… - Expert Systems with …, 2024 - Elsevier
User intents are ever-changing, which requires deep learning models to have the ability to
classify unknown intents. Meta-learning aims to solve this problem by improving the model's …

Few-shot cyberviolence intent classification with Meta-learning AutoEncoder based on adversarial domain adaptation

S Yang, YJ Du, SY Du, XY Li, XL Chen, YL Li, CZ Xie… - Neurocomputing, 2024 - Elsevier
The phenomenon of cyberviolence has become a critical issue in online security, drawing
attention from various stakeholders. A major shortcoming in the previous works is the …

Generative Probabilistic Meta-Learning for Few-Shot Image Classification

M Fu, X Wang, J Wang, Z Yi - IEEE Transactions on Emerging …, 2024 - ieeexplore.ieee.org
Meta-learning, a rapidly advancing area in computational intelligence, leverages prior
knowledge from related tasks to facilitate the swift adaptation to new tasks with limited data …

Few-shot intent detection with mutual information and contrastive learning

S Yang, YJ Du, JM Huang, XY Li, SY Du, J Liu… - Applied Soft …, 2024 - Elsevier
Few-shot intent detection is a challenging task. Most existing methods only focus on
acquisition of generalization knowledge in known classes, or on the adaptation situation of …

Meta-Learning With Versatile Loss Geometries for Fast Adaptation Using Mirror Descent

Y Zhang, B Li, GB Giannakis - ICASSP 2024-2024 IEEE …, 2024 - ieeexplore.ieee.org
Utilizing task-invariant prior knowledge extracted from related tasks, meta-learning is a
principled framework that empowers learning a new task especially when data records are …

Meta-Learning Priors Using Unrolled Proximal Networks

Y Zhang, GB Giannakis - The Twelfth International Conference on Learning … - openreview.net
Relying on prior knowledge accumulated from related tasks, meta-learning offers a powerful
approach to learning a novel task from a limited number of training data. Recent approaches …

Meta-Learning Universal Priors Using Non-Injective Change of Variables

Y Zhang, A Sadeghi, GB Giannakis - The Thirty-eighth Annual Conference … - openreview.net
Meta-learning empowers data-hungry deep neural networks to rapidly learn from merely a
few samples, which is especially appealing to tasks with small datasets. Critical in this …

Meta-Learning Universal Priors Using Non-Injective Normalizing Flows

Y Zhang, A Sadeghi, GB Giannakis - openreview.net
Meta-learning empowers data-hungry deep neural networks to rapidly learn from merely a
few samples, which is especially appealing to tasks with small datasets. Critical in this …