Thomas O'Leary-Roseberry

Inexact Newton Methods for Stochastic Non-Convex Optimization with Applications to Neural Network Training

Thomas O'Leary-Roseberry, Nick Alger, Omar Ghattas, 2019, 1905.06738.

Projected Stein Variational Newton: A Fast and Scalable Bayesian Inference Method in High Dimensions

Peng Chen, Thomas O'Leary-Roseberry, Keyi Wu, 2019, NeurIPS.

Ill-Posedness and Optimization Geometry for Nonlinear Neural Network Training pdf

Thomas O'Leary-Roseberry, Omar Ghattas, O. Ghattas, 2020, ArXiv.

Low Rank Saddle Free Newton: Algorithm and Analysis pdf

Thomas O'Leary-Roseberry, Nick Alger, Omar Ghattas, 2020, ArXiv.

Derivative-Informed Projected Neural Networks for High-Dimensional Parametric Maps Governed by PDEs pdf

Peng Chen, Thomas O'Leary-Roseberry, Omar Ghattas, 2020, Computer Methods in Applied Mechanics and Engineering.

Adaptive Projected Residual Networks for Learning Parametric Maps from Sparse Data pdf

Joaquim R. R. A. Martins, Karen Willcox, Anirban Chaudhuri, 2021, ArXiv.

Derivative-informed projected neural network for large-scale Bayesian optimal experimental design

Thomas O'Leary-Roseberry, Omar Ghattas, Peng Chen, 2022, ArXiv.

Derivative-Informed Projected Neural Networks for High-Dimensional Parametric Maps Governed by PDEs pdf

Peng Chen, Thomas O'Leary-Roseberry, Omar Ghattas, 2022, Computer Methods in Applied Mechanics and Engineering.