Submission Number: 196
Submission ID: 4466
Submission UUID: a3fa6afc-e4ff-4b27-99a3-b1c2c0d365b6
Submission URI: /form/project

Created: Sat, 04/06/2024 - 05:58
Completed: Sat, 04/06/2024 - 06:01
Changed: Tue, 04/22/2025 - 11:06

Remote IP address: 104.28.39.76
Submitted by: Gaurav Khanna
Language: English

Is draft: No
Webform: Project
Waveform Systematics for Black Hole Binary Mergers Models (extension)
CAREERS
{Empty}
ai (271), AI/ML (802), astrophysics (297), conda (227), cuda (222), distributed-computing (92), jupyterhub (214), neural-networks (435)
Complete

Project Leader

Michael Puerrer
{Empty}
{Empty}

Project Personnel

{Empty}
Samuel Clyne
{Empty}

Project Information

This is an extension of project "Waveform Systematics for Black Hole Binary Mergers Models". That project leveraged the ML Dingo code to compute posterior distribution for gravitational wave signals and created a visual map of measures of discrepancies between the posteriors
obtained for different waveform families for the same set of signals. In this extension we aim to generalize the analysis to more generic black hole binaries sources which can undergo precession of the orbital plane and black hole spins.

Following the first project, the student will focus on training more complex neural networks, perform Bayesian inference with the Python-based Dingo code, and extend the visualizations of discrepancies between posterior distribution on URI’s UNITY cluster.

Project Information Subsection

{Empty}
{Empty}
{Empty}
{Empty}
Some hands-on experience
{Empty}
University of Rhode Island -- Center for Computational Research
{Empty}
CR-University of Rhode Island
{Empty}
No
Already behind3Start date is flexible
3
{Empty}
{Empty}
  • Milestone Title: Milestone #1
    Milestone Description: The student will leverage the Dingo code to train networks for precessing binaries and perform inference over a larger grid of signal parameters which now also vary over in-plane black hole spins; launch presentation.
    Completion Date Goal: 2024-04-30
  • Milestone Title: Milestone #2
    Milestone Description: The student will implement simple waveform inner products ("overlaps") between signals and template waveforms in Python and will explore whether these
    overlaps can serve to predict measures of discrepancy between posteriors such as the bias.
    Completion Date Goal: 2024-05-31
  • Milestone Title: Milestone #3
    Milestone Description: The student will bring together the results generated in the study and
    will create visualizations of discrepancy measures over the signal parameter space using
    Python notebooks (matplotlib, plotly); wrap presentation.
    Completion Date Goal: 2024-06-30
{Empty}
{Empty}
{Empty}
{Empty}
{Empty}
{Empty}
{Empty}
{Empty}

Final Report

{Empty}
{Empty}
{Empty}
{Empty}
{Empty}
{Empty}
{Empty}
{Empty}
{Empty}
{Empty}