% Key papers for SGD noise and flat minima (scime atlas, https://atlas.scime.ai). Authors are not included: add them from the paper pages.

@inproceedings{iclr2017_on_28253,
  title = {{On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2017},
  url = {https://openreview.net/forum?id=H1oyRlYgg},
}

@inproceedings{icml2016_train_36016,
  title = {{Train faster, generalize better: Stability of stochastic gradient descent}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2016},
  url = {https://proceedings.mlr.press/v48/hardt16.html},
}

@inproceedings{iclr2018_stochastic_28562,
  title = {{Stochastic gradient descent performs variational inference, converges to limit cycles for deep networks}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2018},
  url = {https://iclr.cc/virtual/2018/poster/152},
}

@inproceedings{aaai2016_preconditioned_105028,
  title = {{Preconditioned Stochastic Gradient Langevin Dynamics for Deep Neural Networks}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2016},
  url = {https://ojs.aaai.org/index.php/AAAI/article/view/10200},
}

@inproceedings{icml2018_the_37041,
  title = {{The Power of Interpolation: Understanding the Effectiveness of SGD in Modern Over-parametrized Learning}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2018},
  url = {https://icml.cc/virtual/2018/poster/1968},
}

@inproceedings{iclr2017_entropy_28204,
  title = {{Entropy-SGD: Biasing Gradient Descent Into Wide Valleys}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2017},
  url = {https://openreview.net/forum?id=B1YfAfcgl},
}

@inproceedings{neurips2019_sgd_58928,
  title = {{SGD on Neural Networks Learns Functions of Increasing Complexity}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2019},
  url = {https://nips.cc/virtual/2019/poster/14513},
}

@inproceedings{icml2019_a_37110,
  title = {{A Tail-Index Analysis of Stochastic Gradient Noise in Deep Neural Networks}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2019},
  url = {https://icml.cc/virtual/2019/poster/3560},
}

@inproceedings{icml2019_the_37790,
  title = {{The Anisotropic Noise in Stochastic Gradient Descent: Its Behavior of Escaping from Sharp Minima and Regularization Effects}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2019},
  url = {https://icml.cc/virtual/2019/poster/3792},
}

@inproceedings{neurips2018_how_57208,
  title = {{How SGD Selects the Global Minima in Over-parameterized Learning: A Dynamical Stability Perspective}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2018},
  url = {https://nips.cc/virtual/2018/poster/11792},
}

@inproceedings{icml2018_data_36613,
  title = {{Data-Dependent Stability of Stochastic Gradient Descent}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2018},
  url = {https://icml.cc/virtual/2018/poster/1984},
}

@inproceedings{neurips2020_towards_60998,
  title = {{Towards Theoretically Understanding Why Sgd Generalizes Better Than Adam in Deep Learning}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2020},
  url = {https://nips.cc/virtual/2020/poster/16875},
}

@inproceedings{iclr2019_fluctuation_29062,
  title = {{Fluctuation-dissipation relations for stochastic gradient descent}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2019},
  url = {https://iclr.cc/virtual/2019/poster/784},
}

@inproceedings{iclr2021_a_30739,
  title = {{A Diffusion Theory For Deep Learning Dynamics: Stochastic Gradient Descent Exponentially Favors Flat Minima}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2021},
  url = {https://iclr.cc/virtual/2021/poster/2620},
}

@inproceedings{icml2018_an_36509,
  title = {{An Alternative View: When Does SGD Escape Local Minima?}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2018},
  url = {https://icml.cc/virtual/2018/poster/1941},
}

@inproceedings{icml2020_on_38555,
  title = {{On the Noisy Gradient Descent that Generalizes as SGD}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2020},
  url = {https://icml.cc/virtual/2020/poster/6307},
}

@inproceedings{iclr2019_sgd_28842,
  title = {{SGD Converges to Global Minimum in Deep Learning via Star-convex Path}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2019},
  url = {https://iclr.cc/virtual/2019/poster/882},
}

@inproceedings{neurips2020_on_60475,
  title = {{On the Almost Sure Convergence of Stochastic Gradient Descent in Non-Convex Problems}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2020},
  url = {https://nips.cc/virtual/2020/poster/18514},
}

@inproceedings{icml2021_the_40017,
  title = {{The Heavy-Tail Phenomenon in SGD}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2021},
  url = {https://icml.cc/virtual/2021/poster/9017},
}

@inproceedings{aaai2021_stability_1415,
  title = {{Stability and Generalization of Decentralized Stochastic Gradient Descent}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2021},
  url = {https://aaai.org/papers/09756-stability-and-generalization-of-decentralized-stochastic-gradient-descent/},
}

@inproceedings{icml2024_improved_44068,
  title = {{Improved Stability and Generalization Guarantees of the Decentralized SGD Algorithm}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2024},
  url = {https://icml.cc/virtual/2024/poster/34384},
}

@inproceedings{iclr2024_stochastic_33875,
  title = {{Stochastic Modified Equations and Dynamics of Dropout Algorithm}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2024},
  url = {https://iclr.cc/virtual/2024/poster/19200},
}

@inproceedings{neurips2024_the_70445,
  title = {{The High Line: Exact Risk and Learning Rate Curves of Stochastic Adaptive Learning Rate Algorithms}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/96661},
}

@inproceedings{neurips2024_emergence_71184,
  title = {{Emergence of heavy tails in homogenized stochastic gradient descent}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/96047},
}

@inproceedings{neurips2024_stochastic_73181,
  title = {{Stochastic Optimization Schemes for Performative Prediction with Nonconvex Loss}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/94252},
}

@inproceedings{neurips2024_parameter_74304,
  title = {{Parameter Symmetry and Noise Equilibrium of Stochastic Gradient Descent}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/93254},
}

@inproceedings{neurips2025_gaussian_78017,
  title = {{Gaussian Approximation and Concentration of Constant Learning-Rate Stochastic Gradient Descent}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=aCPFvEg22L},
}

@inproceedings{aaai2025_error_9009,
  title = {{Error Analysis Affected by Heavy-Tailed Gradients for Non-Convex Pairwise Stochastic Gradient Descent}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2025},
  url = {https://ojs.aaai.org/index.php/AAAI/article/view/33735},
}

@inproceedings{cvpr2025_investigating_26653,
  title = {{Investigating the Role of Weight Decay in Enhancing Nonconvex SGD}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2025},
  url = {https://cvpr.thecvf.com/virtual/2025/poster/32954},
}

@inproceedings{iclr2024_risk_33827,
  title = {{Risk Bounds of Accelerated SGD for Overparameterized Linear Regression}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2024},
  url = {https://iclr.cc/virtual/2024/poster/19248},
}

@inproceedings{iclr2024_butterfly_33912,
  title = {{Butterfly Effects of SGD Noise: Error Amplification in Behavior Cloning and Autoregression}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2024},
  url = {https://iclr.cc/virtual/2024/poster/19163},
}

@inproceedings{iclr2024_a_34546,
  title = {{A Precise Characterization of SGD Stability Using Loss Surface Geometry}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2024},
  url = {https://iclr.cc/virtual/2024/poster/18528},
}

@inproceedings{iclr2024_benign_35582,
  title = {{Benign Oscillation of Stochastic Gradient Descent with Large Learning Rate}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2024},
  url = {https://iclr.cc/virtual/2024/poster/17491},
}

@inproceedings{icml2024_implicit_43793,
  title = {{Implicit Compressibility of Overparametrized Neural Networks Trained with Heavy-Tailed SGD}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2024},
  url = {https://icml.cc/virtual/2024/poster/34659},
}

@inproceedings{icml2024_catapults_45134,
  title = {{Catapults in SGD: spikes in the training loss and their impact on generalization through feature learning}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2024},
  url = {https://icml.cc/virtual/2024/poster/33318},
}

@inproceedings{icml2024_what_45682,
  title = {{What is the Long-Run Distribution of Stochastic Gradient Descent? A Large Deviations Analysis}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2024},
  url = {https://icml.cc/virtual/2024/poster/32770},
}

@inproceedings{icml2025_the_46432,
  title = {{The Global Convergence Time of Stochastic Gradient Descent in Non-Convex Landscapes: Sharp Estimates via Large Deviations}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/46133},
}

@inproceedings{icml2026_momentum_54143,
  title = {{Momentum Further Constrains Sharpness at the Edge of Stochastic Stability}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=mL4i6z7Miy},
}

@inproceedings{neurips2024_sgd_72312,
  title = {{SGD vs GD: Rank Deficiency in Linear Networks}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/95034},
}

@inproceedings{neurips2024_estimating_73848,
  title = {{Estimating Generalization Performance Along the Trajectory of Proximal SGD in Robust Regression}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/93661},
}