Inference for Multi-messenger Astrophysics Workshop

May 30, 2019 - University of California, Berkeley

Multi-Messenger Astrophysics (MMA) is an exciting new field of science that combines traditional astronomy with observations of gravitational waves and high-energy neutrinos. However, the MMA promise can be realized only if sufficient cyberinfrastructure is available to rapidly handle, combine, and analyze the very large-scale distributed data from all the types of astronomical measurements. This workshop, on the UC Berkeley campus on May 30th, is focused on the algorithmic and computational approaches to inference in the MMA era. Please register your interest if you'd like to participate.

The topics of particular interest to this group are:

    • Bayesian approaches to rapid inference (with noisy, possibly censored data)
    • Surrogate modelling for parameter estimation of the physics that give rise to the observed data
    • Large dimensional, rare event statistics
    • Semi/supervised inference on imaging/timeseries with neural networks
    • Fast, unsupervised anomaly detection
    • Compute constrained learning / prediction (e.g., prediction on edge devices)
    • Private/secure machine learning (respecting collaboration boundaries)
    • Novel compute architectures (GPU/TPU, RISC-V/System on a Chip, Purpose-built FPGAs)

Scientific Organizing Committee

  • Joshua Bloom (UC Berkeley)
  • David Hogg (NYU)
  • Peter Nugent (LBL)

Local Organizing Committee

  • Stacey Dorton
  • Stefan van der Walt

We are actively soliciting speakers and registration of interest to attend. Please contact us at workshop@ml4science.org if you have any questions.

Organization Schedule

March 1 - Workshop website live, solicitations sent

April 1 - Preliminary participants list posted

May 1 - Preliminary schedule of events posted

May 15 - Final schedule posted

May 28 - (Evening) Arrival

May 29 - Physics in ML Workshop

May 29 (Evening) - Joint reception for both workshops

May 30 - This Workshop

May 31 - Departure

Travel

Those asking for travel support should hold off on making their own arrangements for now. We will be in touch separately on the travel logistics.

These are the workshop hotels if you are planning on making your own arrangements:

Immediately preceding this Workshop, at the same venue, we will be putting on a related workshop focused on the deeper connections of physics and machine learning. Participants are welcome in both. More...

NSF Logo

Sponsored by the National Science Foundation, “Conceptualization of a Scalable Cyberinfrastructure Center for Multimessenger Astrophysics” (solicitation NSF 18-076)