% Key papers for Neural network approximation theory (scime atlas, https://atlas.scime.ai). Authors are not included: add them from the paper pages.

@inproceedings{icml2017_recovery_36359,
  title = {{Recovery Guarantees for One-hidden-layer Neural Networks}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2017},
  url = {https://icml.cc/virtual/2017/poster/752},
}

@inproceedings{icml2017_depth_36145,
  title = {{Depth-Width Tradeoffs in Approximating Natural Functions with Neural Networks}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2017},
  url = {https://icml.cc/virtual/2017/poster/593},
}

@inproceedings{icml2019_sorting_37747,
  title = {{Sorting Out Lipschitz Function Approximation}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2019},
  url = {https://icml.cc/virtual/2019/poster/3940},
}

@inproceedings{icml2016_l1_35877,
  title = {{L1-regularized Neural Networks are Improperly Learnable in Polynomial Time}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2016},
  url = {https://proceedings.mlr.press/v48/zhangd16.html},
}

@inproceedings{neurips2017_on_56532,
  title = {{On the Complexity of Learning Neural Networks}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2017},
  url = {https://nips.cc/virtual/2017/poster/9326},
}

@inproceedings{neurips2020_exactly_59814,
  title = {{Exactly Computing the Local Lipschitz Constant of ReLU Networks}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2020},
  url = {https://nips.cc/virtual/2020/poster/18019},
}

@inproceedings{neurips2019_on_58711,
  title = {{On Learning Over-parameterized Neural Networks: A Functional Approximation Perspective}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2019},
  url = {https://nips.cc/virtual/2019/poster/13454},
}

@inproceedings{neurips2020_rational_60669,
  title = {{Rational neural networks}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2020},
  url = {https://nips.cc/virtual/2020/poster/16818},
}

@inproceedings{neurips2020_on_60482,
  title = {{On the Modularity of Hypernetworks}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2020},
  url = {https://nips.cc/virtual/2020/poster/17901},
}

@inproceedings{icml2017_neural_36306,
  title = {{Neural Networks and Rational Functions}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2017},
  url = {https://icml.cc/virtual/2017/poster/761},
}

@inproceedings{iclr2021_universal_29925,
  title = {{Universal approximation power of deep residual neural networks via nonlinear control theory}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2021},
  url = {https://iclr.cc/virtual/2021/poster/3010},
}

@inproceedings{cvpr2021_towards_18019,
  title = {{Towards Evaluating and Training Verifiably Robust Neural Networks}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2021},
}

@inproceedings{neurips2017_concentration_56205,
  title = {{Concentration of Multilinear Functions of the Ising Model with Applications to Network Data}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2017},
  url = {https://nips.cc/virtual/2017/poster/8799},
}

@inproceedings{neurips2017_eigenvalue_56291,
  title = {{Eigenvalue Decay Implies Polynomial-Time Learnability for Neural Networks}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2017},
  url = {https://nips.cc/virtual/2017/poster/9007},
}

@inproceedings{neurips2019_tight_59094,
  title = {{Tight Sample Complexity of Learning One-hidden-layer Convolutional Neural Networks}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2019},
  url = {https://nips.cc/virtual/2019/poster/14068},
}

@inproceedings{iclr2020_effect_29841,
  title = {{Effect of Activation Functions on the Training of Overparametrized Neural Nets}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2020},
  url = {https://openreview.net/forum?id=rkgfdeBYvH},
}

@inproceedings{aaai2023_on_4277,
  title = {{On Solution Functions of Optimization: Universal Approximation and Covering Number Bounds}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2023},
  url = {https://ojs.aaai.org/index.php/AAAI/article/view/25981},
}

@inproceedings{neurips2020_a_59295,
  title = {{A new convergent variant of Q-learning with linear function approximation}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2020},
  url = {https://nips.cc/virtual/2020/poster/17057},
}

@inproceedings{iclr2020_a_29828,
  title = {{A closer look at the approximation capabilities of neural networks}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2020},
  url = {https://openreview.net/forum?id=rkevSgrtPr},
}

@inproceedings{neurips2020_zap_61116,
  title = {{Zap Q-Learning With Nonlinear Function Approximation}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2020},
  url = {https://nips.cc/virtual/2020/poster/17500},
}

@inproceedings{aaai2024_complexity_5443,
  title = {{Complexity of Neural Network Training and ETR: Extensions with Effectively Continuous Functions}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2024},
  url = {https://ojs.aaai.org/index.php/AAAI/article/view/29118},
}

@inproceedings{neurips2024_non_72590,
  title = {{Non-asymptotic Approximation Error Bounds of Parameterized Quantum Circuits}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/94787},
}

@inproceedings{iclr2025_generalization_83554,
  title = {{Generalization Bounds and Model Complexity for Kolmogorov–Arnold Networks}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2025},
  url = {https://iclr.cc/virtual/2025/poster/28263},
}

@inproceedings{neurips2025_in_77407,
  title = {{In-context Learning of Linear Dynamical Systems with Transformers: Approximation Bounds and Depth-separation}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=TVD7cVIPCp},
}

@inproceedings{iclr2024_novel_34092,
  title = {{Novel Quadratic Constraints for Extending LipSDP beyond Slope-Restricted Activations}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2024},
  url = {https://iclr.cc/virtual/2024/poster/18983},
}

@inproceedings{icml2025_ehrenfeucht_46309,
  title = {{Ehrenfeucht-Haussler Rank and Chain of Thought}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/46252},
}

@inproceedings{icml2025_universal_46635,
  title = {{Universal Approximation Theorem of Deep Q-Networks}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/45935},
}

@inproceedings{icml2025_exact_47681,
  title = {{Exact Upper and Lower Bounds for the Output Distribution of Neural Networks with Random Inputs}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/44916},
}

@inproceedings{icml2026_a_51255,
  title = {{A Noise Sensitivity Exponent Controls Large Statistical-to-Computational Gaps in Single- and Multi-Index Models}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=KHhO311xZ2},
}

@inproceedings{icml2026_rational_53537,
  title = {{Rational Neural Networks have Expressivity Advantages}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=gPQQyHFbEA},
}

@inproceedings{icml2026_approximation_54162,
  title = {{Approximation Error Upper and Lower Bounds for Hölder Class with Transformers}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=mWBxEAq9lv},
}

@inproceedings{neurips2025_optimal_76156,
  title = {{Optimal Minimum Width for the Universal Approximation of Continuously Differentiable Functions by Deep Narrow MLPs}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=GAa7Gy7FW7},
}

@inproceedings{neurips2025_nonparametric_76333,
  title = {{Nonparametric Quantile Regression with ReLU-Activated Recurrent Neural Networks}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=IGKluZvI6J},
}

@inproceedings{neurips2025_monotone_77205,
  title = {{Monotone and Separable Set Functions: Characterizations and Neural Models}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=RSSk1SHxrH},
}

@inproceedings{neurips2025_from_78476,
  title = {{From Kolmogorov to Cauchy: Shallow XNet Surpasses KANs}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=f6AoMP75dy},
}

@inproceedings{neurips2025_robustly_78555,
  title = {{Robustly Learning Monotone Single-Index Models}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=fuOEekc2t3},
}

@inproceedings{neurips2025_approximation_79681,
  title = {{Approximation theory for 1-Lipschitz ResNets}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=rsTWR7z4PB},
}

@inproceedings{iclr2025_decomposition_83910,
  title = {{Decomposition Polyhedra of Piecewise Linear Functions}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2025},
  url = {https://iclr.cc/virtual/2025/poster/27910},
}

@inproceedings{iclr2025_semialgebraic_84166,
  title = {{Semialgebraic Neural Networks: From roots to representations}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2025},
  url = {https://iclr.cc/virtual/2025/poster/27662},
}

@inproceedings{iclr2026_on_88911,
  title = {{On the Lipschitz Continuity of Set Aggregation Functions and Neural Networks for Sets}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2026},
  url = {https://openreview.net/forum?id=sPRK6XefjY},
}