J. Michael Burgess

Building ML infrastructure for property intelligence using geospatial computer vision and satellite-imagery analysis — extracting property characteristics for insurance, real estate, and risk assessment.

Developed AI cytometry for automated cell-population classification and clinical flow-cytometry analysis; built privacy-first, GDPR-validated, reproducible ML workflows.

Member of Dr. Jochen Greiner's group studying GRB emission physics across optical to high-energy gamma-ray instruments. Helped coordinate the group in project design and student training, implemented an automated burst-alert and localization system, and conceived a new coded-mask imaging method (supervising its PhD implementation) that substantially increased the sensitivity of INTEGRAL/SPI.

In Felix Ryde's GRB group: developed physical models for GRB spectra and Bayesian software to fit Fermi data, investigated spectral evolution and the validity of classic spectral correlations, and designed a novel scheme to fit Type Ia SNe cosmology data.

Developed the ability to fit physical spectral models to GRB data. Participated in daily satellite operations, data monitoring, and the GCN collaboration distributing GRB trigger information; built many international research collaborations.

Dissertation: Discerning the Physical Properties of Gamma-Ray Bursts via Time-Resolved Analysis with Physical Spectral Models. Advisor: Dr. Robert D. Preece.

Advisor: Dr. Robert D. Preece.

Dual degree in Math and Physics.

50+ publications across high-energy astrophysics and astrostatistics. Full list in the PDF version and on Google Scholar.

The Multi-Mission Maximum Likelihood framework (3ML) provides a common high-level interface and model definition, which allows for an easy, coherent and intuitive modeling of sources using all the available data, no matter their origin. At the same time, thanks to its architecture based on plug-ins, 3ML uses under the hood the official software of each instrument, the only one certified and maintained by the collaboration which built the instrument itself. This guarantees that 3ML is always using the best possible methodology to deal with the data of each instrument.

Astromodels is a very flexible framework to define models for likelihood or Bayesian analysis of astrophysical data.

popsynth is a generic population modeling framework designed to generate synthetic populations of astrophysical sources either for experimental design or calibration of statistical models.

pynchrotron is a numerical code that uses the Chang-Cooper method to solve the Fokker-Planck equation for the injection of electrons and their subsequent cooling via synchrotron emission. A 3ML model is provided to allow for spectral fitting to GRB/AGN data.

cosmogrb is a package built upon popsynth to simulate GRBs from luminosity functions and various other distributions. Each GRB can be passed through an instrument's response resulting in data that can be later analyzed (preferably with 3ML). Thus, one can generate catalogs of data from theoretical assumptions and test what these assumptions lead to in terms of observation.

gbmgeometry provides spacecraft geometry tools for the Fermi-GBM satellite. 3D plotting and animation tools are provided to allow for understanding the incident angle of GRB emission.

morgoth is the MPE lead GBM localization, spectral analysis, and cataloging pipeline that provides real-time localizations of GRBs to the follow-up community.

An experimental pipeline package to automatically create spectral analysis catalogs for the Fermi-GBM data.

The backbone of BALROG, a full rewrite of the original GBM response generator. BALROG must generate an instrument response for each proposed position on the sky.

PyIPN is a tool for simulating GRB light curves observed by gamma-ray detectors dispersed throughout the Universe (theoretically, but mostly in the Sol system).

A modern approach to triangulating GRBs via time-delay measurements of multiple instruments. The method utilizes random Fourier features and a full-Bayesian hierarchical model to properly locate GRBs with the appropriate statistical likelihood.

A generic response database generator and localization code for energy dispersed non-imaging detectors such as GBM or POLAR-2.

A simple Python solver for generic Fokker-Planck equations.

Watch: youtube.com/watch?v=ewCDt3vzX2I

Watch: youtu.be/n7vWAVhFXiU?t=2988

Prompt Emission Session Chair

Involved in prompt emission panel discussion

Development of new imaging algorithms for INTEGRAL-SPI.

Population synthesis of extra-galactic magnetars.

Fitting of physical models in time and energy to GRB afterglow light curves.

Application of the Nazgul location algorithm to the entire set of IPN GRBs.

Developing population synthesis of GRB infrared afterglows.

Calibrating the effect of localization errors on GRB spectra.

Development of new methods for GRB triangulation.

Bayesian modeling of x-ray binary data obtained by GROND.

Fitting physical afterglow models to multiwavelength data.

Fitting hierarchical Bayesian correlation models to Fermi-GBM data.

Spectral analysis and correlations in GRB spectra.

Synchrotron and photospheric modeling of Fermi-GBM observed GRBs.