Learning feed-forward one-shot learners
arXiv:1606.05233
Abstract
One-shot learning is usually tackled by using generative models or discriminative embeddings. Discriminative methods based on deep learning, which are very effective in other learning scenarios, are ill-suited for one-shot learning as they need large amounts of training data. In this paper, we propose a method to learn the parameters of a deep model in one shot. We construct the learner as a second deep network, called a learnet, which predicts the parameters of a pupil network from a single exemplar. In this manner we obtain an efficient feed-forward one-shot learner, trained end-to-end by minimizing a one-shot classification objective in a learning to learn formulation. In order to make the construction feasible, we propose a number of factorizations of the parameters of the pupil network. We demonstrate encouraging results by learning characters from single exemplars in Omniglot, and by tracking visual objects from a single initial exemplar in the Visual Object Tracking benchmark.
The first three authors contributed equally, and are listed in alphabetical order
References in corpus (3)
Cited by in corpus (34)
- Recent Advances in Open Set Recognition: A Survey
- FSS-1000: A 1000-Class Dataset for Few-Shot Segmentation
- BSNet: Bi-Similarity Network for Few-shot Fine-grained Image Classification
- Few-Example Object Detection with Model Communication
- Piecewise classifier mappings: Learning fine-grained learners for novel categories with few examples
- A Two-Stage Approach to Few-Shot Learning for Image Recognition
- Distilled Siamese Networks for Visual Tracking
- Deep Meta-Learning: Learning to Learn in the Concept Space
- Few-Shot Segmentation Propagation with Guided Networks
- Fast Online Object Tracking and Segmentation: A Unifying Approach
- Learning to Learn: Meta-Critic Networks for Sample Efficient Learning
- Low-Shot Learning from Imaginary Data
- CANet: Class-Agnostic Segmentation Networks with Iterative Refinement and Attentive Few-Shot Learning
- Memory Matching Networks for One-Shot Image Recognition
- Wider and Deeper, Cheaper and Faster: Tensorized LSTMs for Sequence Learning
- Object Relation Detection Based on One-shot Learning
- TAFE-Net: Task-Aware Feature Embeddings for Low Shot Learning
- Generating Neural Networks with Neural Networks
- Adaptive Parameterization for Neural Dialogue Generation
- Dynamic Compositional Neural Networks over Tree Structure
- Parameter-Efficient Transfer from Sequential Behaviors for User Modeling and Recommendation
- On Infinite-Width Hypernetworks
- Deeper, Broader and Artier Domain Generalization
- One-Shot Learning on Attributed Sequences
- Parameter Prediction for Unseen Deep Architectures
- A Generative Model for Sampling High-Performance and Diverse Weights for Neural Networks
- Single-View 3D Object Reconstruction from Shape Priors in Memory
- Online Convolutional Sparse Coding with Sample-Dependent Dictionary
- One-Shot Segmentation in Clutter
- Learning to generate filters for convolutional neural networks
- Towards Robust Pattern Recognition: A Review
- Adversarially-learned Inference via an Ensemble of Discrete Undirected Graphical Models
- Learning to Compose over Tree Structures via POS Tags
- Meta Deformation Network: Meta Functionals for Shape Correspondence