Liam Hodgkinson

46 papers A* 15A 4Journal 27

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	A Regularization-Sharpness Tradeoff for Linear Interpolators. CoRR Qingyi Hu, Liam Hodgkinson
2025	J	`jnl`	ComFe: An Interpretable Head for Vision Transformers. Trans. Mach. Learn. Res. Evelyn J. Mannix, Liam Hodgkinson, Howard D. Bondell
2025	A*	`conf`	Determinant Estimation under Memory Constraints and Neural Scaling Laws. ICML Siavash Ameli, Chris van der Heide, Liam Hodgkinson, Fred Roosta, Michael W. Mahoney
2025	J	`jnl`	Determinant Estimation under Memory Constraints and Neural Scaling Laws. CoRR Siavash Ameli, Chris van der Heide, Liam Hodgkinson, Fred Roosta, Michael W. Mahoney
2025	A*	`conf`	Models of Heavy-Tailed Mechanistic Universality. ICML Liam Hodgkinson, Zhichao Wang, Michael W. Mahoney
2025	J	`jnl`	Models of Heavy-Tailed Mechanistic Universality. CoRR Liam Hodgkinson, Zhichao Wang, Michael W. Mahoney
2025	J	`jnl`	Preserving Angles Improves Feature Distillation. Trans. Mach. Learn. Res. Evelyn J. Mannix, Liam Hodgkinson, Howard D. Bondell
2025	J	`jnl`	Spectral Estimation with Free Decompression. CoRR Siavash Ameli, Chris van der Heide, Liam Hodgkinson, Michael W. Mahoney
2025	A	`conf`	Temperature Optimization for Bayesian Deep Learning. UAI Kenyon Ng, Chris van der Heide, Liam Hodgkinson, Susan Wei
2025	J	`jnl`	Uncertainty Quantification with the Empirical Neural Tangent Kernel. CoRR Joseph Wilson, Chris van der Heide, Liam Hodgkinson, Fred Roosta
2025	J	`jnl`	Uncertainty-Aware Diagnostics for Physics-Informed Machine Learning. CoRR Mara Daniels, Liam Hodgkinson, Michael W. Mahoney
2024	J	`jnl`	Faithful Label-free Knowledge Distillation. CoRR Evelyn J. Mannix, Liam Hodgkinson, Howard D. Bondell
2024	A*	`conf`	How many classifiers do we need? NeurIPS Hyunsuk Kim, Liam Hodgkinson, Ryan Theisen, Michael W. Mahoney
2024	J	`jnl`	How many classifiers do we need? CoRR Hyunsuk Kim, Liam Hodgkinson, Ryan Theisen, Michael W. Mahoney
2024	J	`jnl`	Temperature Optimization for Bayesian Deep Learning. CoRR Kenyon Ng, Chris van der Heide, Liam Hodgkinson, Susan Wei
2023	A*	`conf`	A Heavy-Tailed Algebra for Probabilistic Programming. NeurIPS Feynman T. Liang, Liam Hodgkinson, Michael W. Mahoney
2023	J	`jnl`	A Heavy-Tailed Algebra for Probabilistic Programming. CoRR Feynman T. Liang, Liam Hodgkinson, Michael W. Mahoney
2023	J	`jnl`	A PAC-Bayesian Perspective on the Interpolating Information Criterion. CoRR Liam Hodgkinson, Christopher van der Heide, Robert Salomone, Fred Roosta, Michael W. Mahoney
2023	A*	`conf`	Generalization Guarantees via Algorithm-dependent Rademacher Complexity. COLT Sarah Sachs, Tim van Erven, Liam Hodgkinson, Rajiv Khanna, Umut Simsekli
2023	J	`jnl`	Generalization Guarantees via Algorithm-dependent Rademacher Complexity. CoRR Sarah Sachs, Tim van Erven, Liam Hodgkinson, Rajiv Khanna, Umut Simsekli
2023	A*	`conf`	Monotonicity and Double Descent in Uncertainty Estimation with Gaussian Processes. ICML Liam Hodgkinson, Christopher van der Heide, Fred Roosta, Michael W. Mahoney
2023	A*	`conf`	Test Accuracy vs. Generalization Gap: Model Selection in NLP without Accessing Training or Testing Data. KDD Yaoqing Yang, Ryan Theisen, Liam Hodgkinson, Joseph E. Gonzalez, Kannan Ramchandran, Charles H. Martin, Michael W. Mahoney
2023	J	`jnl`	The Interpolating Information Criterion for Overparameterized Models. CoRR Liam Hodgkinson, Christopher van der Heide, Robert Salomone, Fred Roosta, Michael W. Mahoney
2023	A*	`conf`	When are ensembles really effective? NeurIPS Ryan Theisen, Hyunsuk Kim, Yaoqing Yang, Liam Hodgkinson, Michael W. Mahoney
2023	J	`jnl`	When are ensembles really effective? CoRR Ryan Theisen, Hyunsuk Kim, Yaoqing Yang, Liam Hodgkinson, Michael W. Mahoney
2022	J	`jnl`	Evaluating natural language processing models with generalization metrics that do not need access to any training or testing data. CoRR Yaoqing Yang, Ryan Theisen, Liam Hodgkinson, Joseph E. Gonzalez, Kannan Ramchandran, Charles H. Martin, Michael W. Mahoney
2022	A*	`conf`	Fat-Tailed Variational Inference with Anisotropic Tail Adaptive Flows. ICML Feynman T. Liang, Michael W. Mahoney, Liam Hodgkinson
2022	J	`jnl`	Fat-Tailed Variational Inference with Anisotropic Tail Adaptive Flows. CoRR Feynman T. Liang, Liam Hodgkinson, Michael W. Mahoney
2022	A*	`conf`	Generalization Bounds using Lower Tail Exponents in Stochastic Optimizers. ICML Liam Hodgkinson, Umut Simsekli, Rajiv Khanna, Michael W. Mahoney
2022	J	`jnl`	Monotonicity and Double Descent in Uncertainty Estimation with Gaussian Processes. CoRR Liam Hodgkinson, Christopher van der Heide, Fred Roosta, Michael W. Mahoney
2021	J	`jnl`	Compressing Deep ODE-Nets using Basis Function Expansions. CoRR Alejandro F. Queiruga, N. Benjamin Erichson, Liam Hodgkinson, Michael W. Mahoney
2021	J	`jnl`	Generalization Properties of Stochastic Optimizers via Trajectory Analysis. CoRR Liam Hodgkinson, Umut Simsekli, Rajiv Khanna, Michael W. Mahoney
2021	A	`conf`	Geometric rates of convergence for kernel-based sampling algorithms. UAI Rajiv Khanna, Liam Hodgkinson, Michael W. Mahoney
2021	J	`jnl`	Implicit Langevin Algorithms for Sampling From Log-concave Densities. J. Mach. Learn. Res. Liam Hodgkinson, Robert Salomone, Fred Roosta
2021	A*	`conf`	Lipschitz Recurrent Neural Networks. ICLR N. Benjamin Erichson, Omri Azencot, Alejandro F. Queiruga, Liam Hodgkinson, Michael W. Mahoney
2021	A*	`conf`	Multiplicative Noise and Heavy Tails in Stochastic Optimization. ICML Liam Hodgkinson, Michael W. Mahoney
2021	A*	`conf`	Noisy Recurrent Neural Networks. NeurIPS Soon Hoe Lim, N. Benjamin Erichson, Liam Hodgkinson, Michael W. Mahoney
2021	J	`jnl`	Noisy Recurrent Neural Networks. CoRR Soon Hoe Lim, N. Benjamin Erichson, Liam Hodgkinson, Michael W. Mahoney
2021	A	`conf`	Shadow Manifold Hamiltonian Monte Carlo. AISTATS Christopher van der Heide, Fred Roosta, Liam Hodgkinson, Dirk P. Kroese
2021	A*	`conf`	Stateful ODE-Nets using Basis Function Expansions. NeurIPS Alejandro F. Queiruga, N. Benjamin Erichson, Liam Hodgkinson, Michael W. Mahoney
2021	A	`conf`	Stochastic continuous normalizing flows: training SDEs as ODEs. UAI Liam Hodgkinson, Christopher van der Heide, Fred Roosta, Michael W. Mahoney
2021	A*	`conf`	Taxonomizing local versus global structure in neural network loss landscapes. NeurIPS Yaoqing Yang, Liam Hodgkinson, Ryan Theisen, Joe Zou, Joseph E. Gonzalez, Kannan Ramchandran, Michael W. Mahoney
2021	J	`jnl`	Taxonomizing local versus global structure in neural network loss landscapes. CoRR Yaoqing Yang, Liam Hodgkinson, Ryan Theisen, Joe Zou, Joseph E. Gonzalez, Kannan Ramchandran, Michael W. Mahoney
2020	J	`jnl`	Multiplicative noise and heavy tails in stochastic optimization. CoRR Liam Hodgkinson, Michael W. Mahoney
2020	J	`jnl`	Stochastic Normalizing Flows. CoRR Liam Hodgkinson, Christopher van der Heide, Fred Roosta, Michael W. Mahoney
2019	J	`jnl`	Implicit Langevin Algorithms for Sampling From Log-concave Densities. CoRR Liam Hodgkinson, Robert Salomone, Fred (Farbod) Roosta