Concept Discovery for Fast Adapatation
Not on the map. Why?
Abstract (arXiv)
The advances in deep learning have enabled machine learning methods to outperform human beings in various areas, but it remains a great challenge for a well-trained model to quickly adapt to a new task. One promising solution to realize this goal is through meta-learning, also known as learning to learn, which has achieved promising results in few-shot learning. However, current approaches are still enormously different from human beings' learning process, especially in the ability to extract structural and transferable knowledge. This drawback makes current meta-learning frameworks non-interpretable and hard to extend to more complex tasks. We tackle this problem by introducing concept discovery to the few-shot learning problem, where we achieve more effective adaptation by meta-learning the structure among the data features, leading to a composite representation of the data. Our proposed method Concept-Based Model-Agnostic Meta-Learning (COMAML) has been shown to achieve consistent improvements in the structured data for both synthesized datasets and real-world datasets.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 35 of its 50 references that are in the map. First posted on arXiv 19 Jan 2023.
Builds on
35 of 50 references.
References · 35
- 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
- 2017Neural Discrete Representation Learningcited by 4,650
- 2017Prototypical Networks for Few-shot Learningcited by 3,001
- 2016Matching Networks for One Shot Learningcited by 2,246
- 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
- 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,139
- 2018On First-Order Meta-Learning Algorithmscited by 1,003
- 2020Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inferencecited by 934
- 2019A Closer Look at Few-shot Classificationcited by 679
- 2018Meta-Learning with Latent Embedding Optimizationcited by 476
- 2019Meta-Learning with Differentiable Convex Optimizationcited by 461
- 2018Dynamic Few-Shot Visual Learning without Forgettingcited by 385
23 more
- 2019Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAMLcited by 359
- 2019Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examplescited by 312
- 2018How to train your MAMLcited by 306
- 2018Meta-Transfer Learning for Few-Shot Learningcited by 306
- 2019Adversarial Attacks on Graph Neural Networks via Meta Learningcited by 261
- 2018Probabilistic Model-Agnostic Meta-Learningcited by 254
- 2019Learning Discrete Structures for Graph Neural Networkscited by 228
- 2019Cross Attention Network for Few-shot Classificationcited by 218
- 2018Bayesian Model-Agnostic Meta-Learningcited by 194
- 2018Recasting Gradient-Based Meta-Learning as Hierarchical Bayescited by 180
- 2017Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representationscited by 149
- 2018One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learningcited by 146
- 2018Meta-Reinforcement Learning of Structured Exploration Strategiescited by 128
- 2018Meta-Learning for Low-Resource Neural Machine Translationcited by 105
- 2020Meta-Learning Symmetries by Reparameterizationcited by 75
- 2019Hierarchically Structured Meta-learningcited by 67
- 2019Graph Few-shot Learning via Knowledge Transfercited by 63
- 2018Learning Compositional Representations for Few-Shot Recognitioncited by 50
- 2020DeepEMD: Differentiable Earth Mover's Distance for Few-Shot Learningcited by 49
- 2020Adaptive Task Sampling for Meta-Learningcited by 29
- 2020Online Structured Meta-learningcited by 27
- 2020Meta Reinforcement Learning with Autonomous Inference of Subtask Dependenciescited by 25
- 2021Few-Shot Learning with Part Discovery and Augmentation from Unlabeled Imagescited by 8
Built on by
Not cited yet.