Locality and compositionality in zero-shot learning
arXiv:1912.12179
Abstract
In this work we study locality and compositionality in the context of learning representations for Zero Shot Learning (ZSL). In order to well-isolate the importance of these properties in learned representations, we impose the additional constraint that, differently from most recent work in ZSL, no pre-training on different datasets (e.g. ImageNet) is performed. The results of our experiments show how locality, in terms of small parts of the input, and compositionality, i.e. how well can the learned representations be expressed as a function of a smaller vocabulary, are both deeply related to generalization and motivate the focus on more local-aware models in future research directions for representation learning.
Published at ICLR 2020
References in corpus (12)
- Auto-Encoding Variational Bayes
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- How transferable are features in deep neural networks?
- Self-Attention Generative Adversarial Networks
- QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension
- What makes ImageNet good for transfer learning?
- Approximating CNNs with Bag-of-local-Features models works surprisingly well on ImageNet
- SCAN: Learning Hierarchical Compositional Visual Concepts
- Modular meta-learning
- Visual Space Optimization for Zero-shot Learning
Cited by in corpus (11)
- Zero-shot Node Classification with Decomposed Graph Prototype Network
- Contrastive Learning with Hard Negative Samples
- Simple and effective localized attribute representations for zero-shot learning
- Goal-Oriented Gaze Estimation for Zero-Shot Learning
- Region Semantically Aligned Network for Zero-Shot Learning
- Cross-Trajectory Representation Learning for Zero-Shot Generalization in RL
- Cross-Modal Information Maximization for Medical Imaging: CMIM
- Contrastive Attraction and Contrastive Repulsion for Representation Learning
- Compositional Fine-Grained Low-Shot Learning
- What Remains of Visual Semantic Embeddings
- Concepts, Properties and an Approach for Compositional Generalization