Meta-Learning Initializations for Image Segmentation
We extend first-order model agnostic meta-learning algorithms (including FOMAML and Reptile) to image segmentation, present a novel neural network architecture built for fast learning which we call EfficientLab, and leverage a formal definition of the test error of meta-learning algorithms to decrease error on out of distribution tasks. We show state of the art results on the FSS-1000 dataset by meta-training EfficientLab with FOMAML and using Bayesian optimization to infer the optimal test-time adaptation routine hyperparameters. We also construct a small benchmark dataset, FP-k, for the empirical study of how meta-learning systems perform in both few- and many-shot settings. On the FP-k dataset, we show that meta-learned initializations provide value for canonical few-shot image segmentation but their performance is quickly matched by conventional transfer learning with performance being equal beyond 10 labeled examples. Our code, meta-learned model, and the FP-k dataset are available at https://github.com/ml4ai/mliis .
Code (1)
Tasks
Bayesian OptimizationFew-Shot Image ClassificationFew-Shot LearningImage ClassificationImage SegmentationMeta-LearningSegmentationSemantic SegmentationTest-time AdaptationTransfer LearningMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Meta-Learning GNN Initializations for Low-Resource Molecular Property Prediction
Building $\textit{in silico}$ models to predict chemical properties and activities is a crucial step in drug discovery. However, limited labeled data often hinders the application of deep learning in this setting. Meanwh…
Drug DiscoveryFew-Shot LearningMeta-LearningMolecular Property Prediction+1MetaInit: Initializing learning by learning to initialize
Deep learning models frequently trade handcrafted features for deep features learned with much less human intervention using gradient descent. While this paradigm has been enormously successful, deep networks are often d…
Meta-LearningMeta-Learning Initializations for Low-Resource Drug Discovery
Building in silico models to predict chemical properties and activities is a crucial step in drug discovery. However, drug discovery projects are often characterized by limited labeled data, hindering the applications of…
Drug DiscoveryFew-Shot LearningMeta-LearningModulating transfer between tasks in gradient-based meta-learning
Learning-to-learn or meta-learning leverages data-driven inductive bias to increase the efficiency of learning on a novel task. This approach encounters difficulty when transfer is not mutually beneficial, for instance, …
DiversityInductive BiasMeta-LearningXB-MAML: Learning Expandable Basis Parameters for Effective Meta-Learning with Wide Task Coverage
Meta-learning, which pursues an effective initialization model, has emerged as a promising approach to handling unseen tasks. However, a limitation remains to be evident when a meta-learner tries to encompass a wide rang…
Inductive BiasMeta-Learning