SOTAVerified

Meta-Learning

Meta-learning is a methodology considered with "learning to learn" machine learning algorithms.

( Image credit: Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks )

Papers

Showing 31413150 of 3569 papers

TitleStatusHype
Simultaneous Perturbation Method for Multi-Task Weight Optimization in One-Shot Meta-LearningCode0
Effective Structured Prompting by Meta-Learning and Representative VerbalizerCode0
Improve Meta-learning for Few-Shot Text Classification with All You Can Acquire from the TasksCode0
Zero-Shot Task TransferCode0
MetaKernel: Learning Variational Random Features with Limited LabelsCode0
Im-Promptu: In-Context Composition from Image PromptsCode0
Multi-view Distillation based on Multi-modal Fusion for Few-shot Action Recognition(CLIP-M^2DF)Code0
Meta Label Correction for Noisy Label LearningCode0
MetAL: Active Semi-Supervised Learning on Graphs via Meta LearningCode0
Bootstrapping Informative Graph Augmentation via A Meta Learning ApproachCode0
Show:102550
← PrevPage 315 of 357Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1MZ+ReconMeta-train success rate97.8Unverified
2MZMeta-train success rate97.6Unverified
3MAMLMeta-test success rate36Unverified
4RL^2Meta-test success rate10Unverified
5DnCMeta-test success rate5.4Unverified
6PEARLMeta-test success rate0Unverified
#ModelMetricClaimedVerifiedStatus
1SoftModuleAverage Success Rate60Unverified
2Multi-task multi-head SACAverage Success Rate35.85Unverified
3DisCorAverage Success Rate26Unverified
4NDPAverage Success Rate11Unverified
#ModelMetricClaimedVerifiedStatus
1MZ+ReconMeta-test success rate (zero-shot)18.5Unverified
2MZMeta-test success rate (zero-shot)17.7Unverified
#ModelMetricClaimedVerifiedStatus
1Metadrop% Test Accuracy95.75Unverified