ICML 2018 accepted papers by research area
623 papers were accepted at ICML 2018 (+188 vs 2017). Here is how they split across 20 research areas, and which topics gained the most ground.
See ICML 2018 in the interactive map
Submitting to ICML? Checklist →
Research areas
| Area | Papers | Share | Change in share vs 2017 |
|---|---|---|---|
| Learning theory and algorithms | 149 | 23.9% | -3.0 pts |
| Probabilistic and scientific modeling | 106 | 17.0% | -0.2 pts |
| Optimization | 86 | 13.8% | steady |
| Reinforcement learning | 61 | 9.8% | +4.0 pts |
| Architectures and efficiency | 59 | 9.5% | -1.3 pts |
| Label-efficient and robust learning | 47 | 7.5% | +1.3 pts |
| Bandits and online learning | 24 | 3.9% | -1.9 pts |
| Image generation | 18 | 2.9% | -0.8 pts |
| Graph learning | 14 | 2.2% | -0.7 pts |
| Adversarial robustness and security | 13 | 2.1% | +0.9 pts |
| Multi-agent systems and games | 12 | 1.9% | +0.5 pts |
| Video | 8 | 1.3% | +0.6 pts |
| Federated learning and privacy | 8 | 1.3% | steady |
| Time series | 7 | 1.1% | -0.3 pts |
| Continual learning | 3 | 0.5% | +0.3 pts |
| Detection and segmentation | 3 | 0.5% | +0.3 pts |
| Large language models | 2 | 0.3% | -0.1 pts |
| Vision-language and multimodal | 1 | 0.2% | +0.2 pts |
| 3D vision | 1 | 0.2% | +0.2 pts |
| Point clouds | 1 | 0.2% | +0.2 pts |
Fastest-growing topics at ICML 2018
Change in share of ICML papers from 2017 to 2018, topics with at least 10 papers.
- Visual and representation RL (Reinforcement learning): 8 → 27 papers
- Policy gradient and actor-critic (Reinforcement learning): 11 → 27 papers
- Fairness (Learning theory and algorithms): 1 → 11 papers
- Distributed and compressed (Optimization): 5 → 16 papers
- Implicit bias and generalization (Optimization): 5 → 15 papers
- Kernels and regression theory (Learning theory and algorithms): 8 → 19 papers
- GANs (Image generation): 6 → 13 papers
- Recognition and long-tail (Label-efficient and robust learning): 16 → 27 papers
- Uncertainty and Gaussian processes (Probabilistic and scientific modeling): 14 → 24 papers
- Prediction and decision losses (Learning theory and algorithms): 21 → 31 papers
More venues
Same year: NeurIPS 2018 · ICLR 2018 · CVPR 2018 · AAAI 2018 · ECCV 2018 · TPAMI 2018 · IJCV 2018 · JMLR 2018
ICML in other years: 2016 · 2017 · 2019 · 2020 · 2021 · 2022 · 2023 · 2024 · 2025 · 2026
