 |
 Sophia: A Scalable Stochastic Second-order Optimizer for Language Model
Pre-trainingInternational Conference on Learning Representations (ICLR), 2023 |
 Sketchy: Memory-efficient Adaptive Regularization with Frequent
DirectionsNeural Information Processing Systems (NeurIPS), 2023 |
 Generalisation under gradient descent via deterministic PAC-BayesInternational Conference on Algorithmic Learning Theory (ALT), 2022 |
 Beyond accuracy: generalization properties of bio-plausible temporal
credit assignment rulesNeural Information Processing Systems (NeurIPS), 2022 |
 Neuronal diversity can improve machine learning for physics and beyondScientific Reports (Sci Rep), 2022 |
 When Do Flat Minima Optimizers Work?Neural Information Processing Systems (NeurIPS), 2022 |
 Hessian Eigenspectra of More Realistic Nonlinear ModelsNeural Information Processing Systems (NeurIPS), 2021 |
