ICLR 2019 accepted papers on Label-efficient and robust learning
56 of the 502 papers accepted at ICLR 2019 (11.2%) are on Label-efficient and robust learning, compared with 11.0% at ICLR 2018. They fall into 6 topics.
See ICLR 2019 in the interactive map
Submitting to ICLR? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Recognition and long-tail · 34 papers
- Learning deep representations by mutual information estimation and maximization (1,723 citations)
- Learning protein sequence embeddings using information from structure (219 citations)
- Predicting the Generalization Gap in Deep Networks with Margin Distributions (120 citations)
- Sample Efficient Adaptive Text-to-Speech (104 citations)
- Meta-Learning Update Rules for Unsupervised Representation Learning (86 citations)
- INVASE: Instance-wise Variable Selection using Neural Networks (81 citations)
- Phase-Aware Speech Enhancement with Deep Complex U-Net (77 citations)
- No Training Required: Exploring Random Encoders for Sentence Classification (75 citations)
- Neural Program Repair by Jointly Learning to Localize and Repair (71 citations)
- Multilingual Neural Machine Translation With Soft Decoupled Encoding (66 citations)
- Modeling Uncertainty with Hedged Instance Embeddings (55 citations)
- On the Minimal Supervision for Training Any Binary Classifier from Only Unlabeled Data (49 citations)
and 22 more in the interactive map.
Domain adaptation and generalization · 11 papers
- Learning Robust Representations by Projecting Superficial Statistics Out (90 citations)
- Regularized Learning for Domain Adaptation under Label Shifts (50 citations)
- Unsupervised Domain Adaptation for Distance Metric Learning (46 citations)
- Augmented Cyclic Adversarial Learning for Low Resource Domain Adaptation (44 citations)
- Exemplar Guided Unsupervised Image-to-Image Translation with Semantic Consistency (36 citations)
- Adversarial Domain Adaptation for Stable Brain-Machine Interfaces (27 citations)
- SPIGAN: Privileged Adversarial Learning from Simulation (26 citations)
- Multi-Domain Adversarial Learning (22 citations)
- Emerging Disentanglement in Auto-Encoder Based Unsupervised Image Content Transfer (18 citations)
- Transfer Learning for Sequences via Learning to Collocate (16 citations)
- LEARNING FACTORIZED REPRESENTATIONS FOR OPEN-SET DOMAIN ADAPTATION (12 citations)
Zero-/few-shot and meta-learning · 7 papers
- LEARNING TO PROPAGATE LABELS: TRANSDUCTIVE PROPAGATION NETWORK FOR FEW-SHOT LEARNING (517 citations)
- A Closer Look at Few-shot Classification (185 citations)
- Unsupervised Learning via Meta-Learning (164 citations)
- Learning to Learn with Conditional Class Dependencies (55 citations)
- Transferring Knowledge across Learning Processes (38 citations)
- Adaptive Posterior Learning: few-shot learning with a surprise-based memory module (10 citations)
- Dimensionality Reduction for Representing the Knowledge of Probabilistic Models (8 citations)
Knowledge distillation · 2 papers
- Multilingual Neural Machine Translation with Knowledge Distillation (127 citations)
- Knowledge Flow: Improve Upon Your Teachers (17 citations)
OOD and anomaly detection · 1 paper
- Deep Anomaly Detection with Outlier Exposure (385 citations)
