ICML 2017 accepted papers on Label-efficient and robust learning
27 of the 435 papers accepted at ICML 2017 (6.2%) are on Label-efficient and robust learning, compared with 6.8% at ICML 2016. They fall into 6 topics.
See ICML 2017 in the interactive map
Submitting to ICML? 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 · 16 papers
- Unsupervised Learning by Predicting Noise (208 citations)
- A Unified View of Multi-Label Performance Measures (135 citations)
- Learning the Structure of Generative Models without Labeled Data (110 citations)
- Attentive Recurrent Comparators (77 citations)
- Cognitive Psychology for Deep Neural Networks: A Shape Bias Case Study (61 citations)
- Learning Continuous Semantic Representations of Symbolic Expressions (43 citations)
- Deep Value Networks Learn to Evaluate and Iteratively Refine Structured Outputs (41 citations)
- Risk Bounds for Transferring Representations With and Without Fine-Tuning (34 citations)
- Differentiable Programs with Neural Libraries (28 citations)
- Confident Multiple Choice Learning (26 citations)
- Scalable Generative Models for Multi-label Learning with Missing Labels (21 citations)
- Multi-Class Optimal Margin Distribution Machine (20 citations)
and 4 more in the interactive map.
Domain adaptation and generalization · 4 papers
- Learning to Discover Cross-Domain Relations with Generative Adversarial Networks (726 citations)
- Asymmetric Tri-training for Unsupervised Domain Adaptation (451 citations)
- Deep Transfer Learning with Joint Adaptation Networks (114 citations)
- Source-Target Similarity Modelings for Multi-Source Transfer Gaussian Process Regression (22 citations)
Noisy labels · 2 papers
- Active Heteroscedastic Regression (9 citations)
- Learning to Aggregate Ordinal Labels by Maximizing Separating Width (6 citations)
Zero-/few-shot and meta-learning · 2 papers
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks (6,288 citations)
- Meta Networks (211 citations)
Semi-supervised learning · 2 papers
- Self-Paced Co-training (72 citations)
- Semi-Supervised Classification Based on Classification from Positive and Unlabeled Data (40 citations)
