ICML 2019 accepted papers by research area
774 papers were accepted at ICML 2019 (+151 vs 2018). Here is how they split across 19 research areas, and which topics gained the most ground.
See ICML 2019 in the interactive map
Submitting to ICML? Checklist →
Research areas
| Area | Papers | Share | Change in share vs 2018 |
|---|---|---|---|
| Learning theory and algorithms | 146 | 18.9% | -5.1 pts |
| Probabilistic and scientific modeling | 128 | 16.5% | -0.5 pts |
| Optimization | 104 | 13.4% | -0.4 pts |
| Reinforcement learning | 77 | 9.9% | +0.2 pts |
| Label-efficient and robust learning | 65 | 8.4% | +0.9 pts |
| Architectures and efficiency | 56 | 7.2% | -2.2 pts |
| Bandits and online learning | 41 | 5.3% | +1.4 pts |
| Adversarial robustness and security | 40 | 5.2% | +3.1 pts |
| Graph learning | 36 | 4.7% | +2.4 pts |
| Image generation | 20 | 2.6% | -0.3 pts |
| Federated learning and privacy | 18 | 2.3% | +1.0 pts |
| Multi-agent systems and games | 17 | 2.2% | +0.3 pts |
| Large language models | 9 | 1.2% | +0.8 pts |
| Detection and segmentation | 4 | 0.5% | steady |
| Time series | 4 | 0.5% | -0.6 pts |
| Continual learning | 3 | 0.4% | steady |
| 3D vision | 3 | 0.4% | +0.2 pts |
| Vision-language and multimodal | 2 | 0.3% | steady |
| Video | 1 | 0.1% | -1.2 pts |
Fastest-growing topics at ICML 2019
Change in share of ICML papers from 2018 to 2019, topics with at least 10 papers.
- Certified and empirical robustness (Adversarial robustness and security): 9 → 26 papers
- Online convex optimization and games (Bandits and online learning): 7 → 21 papers
- Variational and generative sampling (Probabilistic and scientific modeling): 41 → 62 papers
- Optimizers (Optimization): 16 → 30 papers
- GNN architectures (Graph learning): 1 → 10 papers
- Differential privacy (Federated learning and privacy): 6 → 12 papers
- Policy gradient and actor-critic (Reinforcement learning): 27 → 38 papers
- Graph algorithms and clustering (Graph learning): 7 → 12 papers
- Implicit bias and generalization (Optimization): 15 → 21 papers
- Recognition and long-tail (Label-efficient and robust learning): 27 → 35 papers
More venues
Same year: NeurIPS 2019 · ICLR 2019 · CVPR 2019 · AAAI 2019 · ICCV 2019 · TPAMI 2019 · IJCV 2019 · JMLR 2019
ICML in other years: 2016 · 2017 · 2018 · 2020 · 2021 · 2022 · 2023 · 2024 · 2025 · 2026
