Meta-Learning-Study

Deepest Season 6 Meta-Learning study papers plus alpha

Those who are new to meta-learning, I recommend to start with reading these

• Model-agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• ICML 2019 Meta-Learning Tutorial [link]
• CS 330: Deep Multi-Task and Meta Learning [link]

Optimization-based Meta-Learning

• Model-agnostic Meta-Learning for Fast Adaptation of Deep Networks, (ICML 2017), [link]
• Meta-Learning with Latent Embedding Optimization, (ICLR 2019), [link]
• How to Train Your MAML, (ICLR 2019), [link]
• Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAML, (NeurIPs 2019 workshop)[link]
• Meta-Learning with Implicit Gradients, (NIPS 2019), [link]
• Meta-Learning with Warped Gradient Descent, (ICLR 2020), [link]

Metric-Learning based Meta-Learning

• Prototypical Networks for Few-shot Learning, (NIPS 2017), [link]
• Learning to Compare: Relation Network for Few-Shot Learning, (CVPR 2018), [link]
• TADAM: Task dependent adaptive metric for improved few-shot learning, (NIPS 2018)[link]
• Infinite Mixture Prototypes for Few-Shot Learning, (ICML 2019), [link]

Black-box adaptation based Meta-Learning

• One-shot Learning with Memory-Augmented Neural Networks, (ArXiv 2016), [link]
• Learning to learn by gradient descent by gradient descent, (NIPS 2016), [link]
• A Simple Neural Attentive Meta-Learner, (ICLR 2018), [link]
• Meta-Learning with Differentiable Convex Optimization, (CVPR 2019), [link]

Bayesian Approaches

• Towards a Neural Statistician, (ICLR 2017), [link]
• Conditional Neural Processes, (ICML 2018), [link]
• Probabilistic Model-Agnostic Meta-Learning, (NIPS 2018), [link]

Generation

• Few-Shot Adversarial Learning of Realistic Neural Talking Head Models, (ICCV 2019), [link]
• Few-Shot Adaptive Gaze Estimation, (ICCV 2019), [link]
• MarioNETte: Few-shot Face Reenactment Preserving Identity of Unseen Targets, (AAAI 2020), [link]
• MetaPix: Few-Shot Video Retargeting, (ICLR 2020), [link]

Unsupervised, Representation

• Unsupervised Learning via Meta-Learning, (ICLR 2019), [link]
• Meta-Learning Update Rules for Unsupervised Representation Learning, (ICLR 2019), [link]

Realistic Setting

• A Closer Look at Few-shot Classification, (ICLR 2019), [link]
• Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples, (ICLR 2020 under review), [link]
• Meta-Learning without Memorization, (ICLR2020), [link]

Object Detection and Segmentation

• CANet: Class-Agnostic Segmentation Networks with Iterative Refinement and Attentive Few-Shot Learning, (CVPR 2019), [link]
• Few-shot Object Detection via Feature Reweighting, (ICCV 2019), [link]
• Meta-Learning to Detect Rare Objects, (ICCV 2019), [link]

Self-Supervised Learning

• Boosting Few-Shot Visual Learning with Self-Supervision, (ICCV 2019), [link]
• Charting the Right Manifold: Manifold Mixup for Few-shot Learning, (ArXiv 2019), [link]

Before 2016