Brian Groenke

brian.groenke@awi.de

About me

I am a Doctoral Researcher at the Alfred Wegener Institute in Potsdam, Germany. My research interests are centered around the application of machine learning in scientific data analysis. Currently, I am working on the integration of statistical estimators into the CryoGrid permafrost model via universal differential equations. Before that, I worked on the application generative learning methods to generate fine scale realizations of climate and weather variables from coarse scale model outputs.

My industry background consists primarily of software engineering, both front-end and back-end development, with a wide range of frameworks and languages.

When I'm not coding or thinking about data, I also enjoy travelling, playing guitar, drinking good beer, and speaking barely passable German.

Education

Ph.D. Computer Science, Technical University of Berlin/Alfred Wegener Institute, Potsdam, 2020 - present

M.Sc. Computer Science, University of Colorado, Boulder, 2018 - 2020

B.Sc. Computer Science and Engineering, The Ohio State University, 2014 - 2018

Community involvement

Sensitivity of Permafrost in the Arctic (SPARC) group, lead by Prof. Dr. Julia Boike, 2020-present

Telegrafenberg ML club, 2020-present

Machine learning research lab, lead by Prof. Claire Monteleoni, 2018-2020

Peer reviewer: Climate Informatics 2019, ICLR 2019, 2020, 2021

Ad hoc reviewer: ICML 2020

Teaching

Teaching Assistant, CSCI 3104, Spring 2020

Graduate Student Staff, CSCI 5622, Fall 2019

Past projects

ClimAlign: Unsupervised statistical downscaling of climate variables via normalizing flows, Fall 2019-Spring 2020

  • Thesis research project supervised by Prof. Claire Monteleoni (Accepted on May 4th, 2020)

  • Developed novel, unsupervised and generative method for statistical downscaling of climate data

Representation learning for coupled climate models, Spring 2019

  • Investigated the application of various dimensionality reduction algorithms and unsupervised learning methods to data from CMIP5

Survey of machine learning methods for climate informatics datasets, Fall 2018

  • Developed and used techniques for evaluating a variety of dimensionality reduction algorithms on climate related RAMPs [1] as part of collaborative course project

AutoMoL: Automated Monadic Logic, 2017-2018

  • Undergraduate student research project under Dr. Neil Tennant, OSU Dept of Philosophy

  • Implemented automated theorem prover and visualization software for the study of proofs in Core Logic (Tennant 2017)

  • Experimented with applying reinforcement learning to automated proof search

NuGRUV: Music generation using deep learning, Fall 2017

  • Modified existing architecture to improve quality and novelty of generated music tracks

Industry Experience

Data Science Intern at Jupiter Intelligence Inc, Boulder, CO, Summer 2019

  • Researched and evaluated the application of deep neural networks for image super resolution to statistical downscaling of low resolution climate model output

Software Development Engineer I at Lake Shore Cryotronics, Westerville, OH, Aug 2017 - May 2019

  • Developed user facing software for Lake Shore’s new M41/M71, M91 FastHall™ instruments as part of core product development team

  • Researched applications of machine learning in sensor manufacturing

Software Engineer Co-op at Lutron Electronics, Coopersburg, PA, Summer 2017

  • Implemented new embedded device firmware update system for Lutron VIVE devices

  • Created a cross-platform mobile app frontend alongside embedded server backend

  • Worked directly on-site with customers on upgrading device firmware

Software Engineer Intern at Lake Shore Cryotronics, Westerville, OH, Jan 2016 - Apr 2017

  • Designed and implemented software/firmware update framework on Android and embedded systems

  • Designed and developed user interfaces for Lake Shore’s Model 155 Voltage/Current Source

Software Validation Engineer Intern at Intel Corporation, Chicago, IL, Summer 2015

  • Implemented system for running and maintaining unit tests in Java codebase for Intel’s Express Service Gateway

  • Implemented framework for gathering and publishing code coverage data from unit/integration testing

Past presentation materials

MS_Thesis__Defense_presentation.pdf
AlignFlow__Cycle_Consistent_Learning_from_Multiple_Domains_via_Normalizing_Flows.pdf
Sylvester_Normalizing_Flows.pdf
Bayesian_GAN.pdf
MOSES_A_Streaming_Alg_For_Lin_DR_2018.pdf
CSCI_7000_NAFCP_Dasgupta_2017.pdf
ML Group 11/9: t-SNE