Talks and presentations
See a map of all the places I've given a talk!
November 02, 2022
Invited Presentations, Institute for Mathematical and Statistical Innovation Workshop: Machine Learning for Climate and Weather Applications, Chicago, IL
This talk introduces a sampling strategy designed to overcome set imbalance in high dimensional datasets in regression tasks. In a case study of training emulators of a gravity wave parameterization scheme on a long-tail distributed dataset, we find that this strategy improves the errors at the tail of the distribution except at the extreme end, while maintaining minimal loss of accuracy at the peak of the distribution.
April 29, 2022
Contributed Presentations, Stratospheric Processes and its Role in Climate (SPARC) Gravity Wave Symposium, Frankfurt am Main, Germany
I talked about some challenges of training machine learning emulators of an existing gravity wave parameterization scheme at the 2021 SPARC Gravity Wave symposium.
December 03, 2021
Contributed Presentations, 2021 International Symposium on Data Assimilation- Online: Machine Learning for Data Assimilation, online
This presentation was a part of the Machine Learning for Data Assimilation portion of the ISDA 2021 events.
October 29, 2021
Invited Presentations, Atmosphere Ocean Science Colloquium, Courant Institute of Mathematical Sciences, New York University, New York, NY
I presented the last chapter of my dissertation to the CAOS seminar at Courant. It was a great chance to meet the CAOS faculty, postdocs, and students as a newcomer to the department!
March 17, 2020
Contributed Presentations, Geophysical and Astrophysical Fluid Dynamics Seminar, CU Boulder, Boulder, Colorado
In this talk, Ian Grooms introduced background on wave turbulence, and then I presented my research on methods for time integration of wave turbulence problems using IMEX and exponential integrators. This work led to this paper.
March 01, 2019
Invited Presentations, Minisymposium: Mitigating Communication Costs Using Variable Precision Computing Techniques @ SIAM Computational Science and Engineering, Spokane, WA
This presentation discussed round off error analysis of linear algebra subroutines used for the Householder QR decomposition, as well as its implementation for use in graph analysis task. Further work on this topic resulted in this paper.
