Consistent Meta-Regularization for Better Meta-Knowledge in Few-Shot Learning

Tian, Pinzhuo; Li, Wenbin

doi:10.1109/tnnls.2021.3084733

Cited by 31 publications

(6 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To evaluate the effectiveness of TPFSL, we compared its performance against seven different benchmarks, including MCFSL [30], SUFSL [46], MAMIFSL [47], DCRFSL [35], IFSL [48], MK-FSL [49], and ProtoNet [50].…”

Section: Methodsmentioning

confidence: 99%

“…ProtoNet [50] 65.60 ± 0.98 83.60 ± 0.68 IFSL [48] 69.84 ± 1.01 84.09 ± 0.68 MAMIFSL [47] 69.99 ± 0.99 84.44 ± 0.67 MKFSL [49] 79.15 ± 0.99 89.23 ± 0.68 DCRFSL [35] 79.17 ± 0.98 89.76 ± 0.68 MCFSL [30] 79.76 ± 0.99 89.05 ± 0.67 SUFSL [46] 76 [48] 67.88 ± 1.07 84.28 ± 0.87 MAMIFSL [47] 68.33 ± 1.09 84.81 ± 0.87 MKFSL [49] 68.15 ± 1.08 84.12 ± 0.87 DCRFSL [35] 68.78 ± 1.08 84.83 ± 0.87 MCFSL [30] 67.69 ± 1.07 83.68 ± 0.87 SUFSL [46] 64. 28 [48] 64.94 ± 1.00 81.02 ± 0.76 MAMIFSL [47] 64.77 ± 1.00 80.12 ± 0.76 MKFSL [49] 65.07 ± 1.00 81.26 ± 0.76 DCRFSL [35] 64.98 ± 1.00 80.46 ± 0.76 MCFSL [30] 64.78 ± 1.00 80.87 ± 0.76 SUFSL [46] 67.07 ± 1.00 87. -Euclidean distance: Euclidean distance is a prevalent distance metric utilized in FSL.…”

Section: Sarscov2mentioning

confidence: 99%

See 1 more Smart Citation

A Few-Shot Learning Approach for Covid-19 Diagnosis Using Quasi-Configured Topological Spaces

Liu,

Wang,

Jiang

et al. 2023

Journal of Artificial Intelligence and Soft Computing Research

View full text Add to dashboard Cite

Accurate and efficient COVID-19 diagnosis is crucial in clinical settings. However, the limited availability of labeled data poses a challenge for traditional machine learning algorithms. To address this issue, we propose Turning Point (TP), a few-shot learning (FSL) approach that leverages high-level turning point mappings to build sophisticated representations across previously labeled data. Unlike existing FSL models, TP learns using quasi-configured topological spaces and efficiently combines the outputs of diverse TP learners. We evaluated TPFSL using three COVID-19 datasets and compared it with seven different benchmarks. Results show that TPFSL outperformed the top-performing benchmark models in both one-shot and five-shot tasks, with an average improvement of 4.50% and 4.43%, respectively. Additionally, TPFSL significantly outperformed the ProtoNet benchmark by 12.966% and 11.033% in one-shot and five-shot classification problems across all datasets. Ablation experiments were also conducted to analyze the impact of variables such as TP density, network topology, distance measure, and TP placement. Overall, TPFSL has the potential to improve the accuracy and speed of diagnoses for COVID-19 in clinical settings and can be a valuable tool for medical professionals.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Sarscov2mentioning

confidence: 99%

A Few-Shot Learning Approach for Covid-19 Diagnosis Using Quasi-Configured Topological Spaces

Liu,

Wang,

Jiang

et al. 2023

Journal of Artificial Intelligence and Soft Computing Research

View full text Add to dashboard Cite

show abstract

“…Moreover, a concept discriminator was designed to recognize different images. Tian et al [125] proposed a new consistent meta-regularization (Con-MetaReg) to enhance the learning ability of meta-learning models. Specifically, a base learner trained on the support set, then another learner trained on a novel query set.…”

Section: Optimization-based Task-specific Feature Representation Lear...mentioning

confidence: 99%

Few-Shot Fine-Grained Image Classification: A Comprehensive Review

Ren,

Li,

et al. 2024

View full text Add to dashboard Cite

Few-shot fine-grained image classification (FSFGIC) methods refer to the classification of images (e.g., birds, flowers, and airplanes) belonging to different subclasses of the same species by a small number of labeled samples. Through feature representation learning, FSFGIC methods can make better use of limited sample information, learn more discriminative feature representations, greatly improve the classification accuracy and generalization ability, and thus achieve better results in FSFGIC tasks. In this paper, starting from the definition of FSFGIC, a taxonomy of feature representation learning for FSFGIC is proposed. According to this taxonomy, we discuss key issues on FSFGIC (including data augmentation, local and/or global deep feature representation learning, class representation learning, and task-specific feature representation learning). In addition, the existing popular datasets, current challenges and future development trends of feature representation learning on FSFGIC are also described.

show abstract

“…While ANIL offers a streamlined architecture and inherent computational advantages, it faces significant challenges that hinder its flexibility and generalizability in the context of meta-learning. These obstacles may include limitations in handling diverse and complex data types, difficulties in adapting to new and unseen scenarios with fewer data, and challenges in effectively transferring knowledge across different domains [5].…”

Section: Introductionmentioning

confidence: 99%

LaANIL: ANIL with Look-Ahead Meta-Optimization and Data Parallelism

Tammisetti,

Bierzynski,

Stettinger

et al. 2024

Electronics

View full text Add to dashboard Cite

Meta-few-shot learning algorithms, such as Model-Agnostic Meta-Learning (MAML) and Almost No Inner Loop (ANIL), enable machines to learn complex tasks quickly with limited data and based on previous experience. By maintaining the inner loop head of the neural network, ANIL leads to simpler computations and reduces the complexity of MAML. Despite its benefits, ANIL suffers from issues like accuracy variance, slow initial learning, and overfitting, hardening its adaptation and generalization. This work proposes “Look-Ahead ANIL” (LaANIL), an enhancement to ANIL for better learning. LaANIL reorganizes ANIL’s internal architecture, integrating parallel computing techniques (to process multiple training examples simultaneously across computing units) and incorporating Nesterov momentum (which accelerates convergence by adjusting the learning rate based on past gradient information and extracting informative features for look-ahead gradient computation). These additional features make our model more state-of-the-art capable and better edge-compatible and thus improve few-short learning by enabling models to quickly adapt to new information and tasks. LaANIL’s effectiveness is validated on established meta-few-shot learning datasets, including FC100, CIFAR-FS, Mini-ImageNet, CUBirds-200-2011, and Tiered-ImageNet. The proposed model achieved an increased validation accuracy by 7 ± 0.7% and a variance reduction by 44 ± 4% in two-way two-shot classification as well as increased validation by 5 ± 0.4% and a variance reduction by 18 ± 2% in five-way five-shot classification on the FC100 dataset and similarly performed well on other datasets.

show abstract

Consistent Meta-Regularization for Better Meta-Knowledge in Few-Shot Learning

Cited by 31 publications

References 20 publications

A Few-Shot Learning Approach for Covid-19 Diagnosis Using Quasi-Configured Topological Spaces

A Few-Shot Learning Approach for Covid-19 Diagnosis Using Quasi-Configured Topological Spaces

Few-Shot Fine-Grained Image Classification: A Comprehensive Review

LaANIL: ANIL with Look-Ahead Meta-Optimization and Data Parallelism

Contact Info

Product

Resources

About