The Neural Covariance SDE: Shaped Infinite Depth-and-Width Networks at Initialization

von Li, Mufan
Veröffentlicht in NeurIPS (36. : 2022 : New Orleans, La.; Online) 36th Conference on Neural Information Processing Systems (NeurIPS 2022 ; Volume 15 of 50 (2023)

Adaptively Exploiting d-Separators with Causal Bandits

von Bilodeau, Blair
Veröffentlicht in NeurIPS (36. : 2022 : New Orleans, La.; Online) 36th Conference on Neural Information Processing Systems (NeurIPS 2022 ; Volume 27 of 50 (2023)

The future is log-Gaussian: ResNets and their infinite-depth-and-width limit at initialization

von Li, Mufan
Veröffentlicht in NeurIPS (35. : 2021 : Online) 35th Conference on Neural Information Processing Systems (NeurIPS 2021) ; Volume 10 of 36 (2022)

Towards a Unified Information-Theoretic Framework for Generalization

von Haghifam, Mahdi
Veröffentlicht in NeurIPS (35. : 2021 : Online) 35th Conference on Neural Information Processing Systems (NeurIPS 2021) ; Volume 32 of 36 (2022)

Minimax Optimal Quantile and Semi-Adversarial Regret via Root-Logarithmic Regularizers

von Negrea, Jeffrey
Veröffentlicht in NeurIPS (35. : 2021 : Online) 35th Conference on Neural Information Processing Systems (NeurIPS 2021) ; Volume 31 of 36 (2022)

Pruning’s Effect on Generalization Through the Lens of Training and Regularization

von Jin, Tian
Veröffentlicht in NeurIPS (36. : 2022 : New Orleans, La.; Online) 36th Conference on Neural Information Processing Systems (NeurIPS 2022 ; Volume 49 of 50 (2023)