 Analytic theory of dropout regularizationPhysical Review E (Phys. Rev. E), 2025 |
 Hadamard product in deep learning: Introduction, Advances and ChallengesIEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025 |
 Dropout Drops Double DescentJapanese Journal of Statistics and Data Science (JSDS), 2023 |
 UMIX: Improving Importance Weighting for Subpopulation Shift via
Uncertainty-Aware MixupNeural Information Processing Systems (NeurIPS), 2022 |
 A Unified Analysis of Mixed Sample Data Augmentation: A Loss Function
PerspectiveNeural Information Processing Systems (NeurIPS), 2022 |
 Implicit regularization of dropoutIEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2022 |
 Boosting Factorization Machines via Saliency-Guided MixupIEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2022 |
 Gating Dropout: Communication-efficient Regularization for Sparsely
Activated TransformersInternational Conference on Machine Learning (ICML), 2022 |
 Exact Solutions of a Deep Linear NetworkNeural Information Processing Systems (NeurIPS), 2022 |
 The Flip Side of the Reweighted Coin: Duality of Adaptive Dropout and
RegularizationNeural Information Processing Systems (NeurIPS), 2021 |
 What training reveals about neural network complexityNeural Information Processing Systems (NeurIPS), 2021 |
 Meta-Learning with Fewer Tasks through Task InterpolationInternational Conference on Learning Representations (ICLR), 2021 |
 Noisy Recurrent Neural NetworksNeural Information Processing Systems (NeurIPS), 2021 |
 On Convergence and Generalization of Dropout TrainingNeural Information Processing Systems (NeurIPS), 2020 |
 How Does Mixup Help With Robustness and Generalization?International Conference on Learning Representations (ICLR), 2020 |
 Explicit Regularisation in Gaussian Noise InjectionsNeural Information Processing Systems (NeurIPS), 2020 |
 The Implicit and Explicit Regularization Effects of DropoutInternational Conference on Machine Learning (ICML), 2020 |
 Implicit Regularization and Convergence for Weight NormalizationNeural Information Processing Systems (NeurIPS), 2019 |
