In-storage computing for multi-messenger astronomy in neutrino experiments and cosmological surveys

Overview

Development stage

In development or acquisitionSource: a) Pre-deployment – The use case is in a development or acquisition status.

High-impact designation

Not high-impactSource: c) Not high-impact

Impact justification

The use case does not have an effect on civil rights/liberties/privacy, access to education/housing/insurance/credit/employment, access to critical government resources/services, human health/safety, critical infrastructure/public safety

Start date

Not reported

Withheld from public reporting

NoSource: a) No

Mission

Topic area

Science

Operational functions

No function tag matched Derived by keyword rules; see methodology

Problem the AI is intended to solve

This project aims to address the big-data challenges and stringent time constraints facing multi-messenger astronomy (MMA) in neutrino experiments and cosomological surveys. Instead of following the traditional computing paradigm of moving data to th

Expected benefits

The purpose is to enhance the ability of large scale neutrino experiments like DUNE to detect neutrinos from core-collapse supernovas (CCSNs) and to extract useful information about their source in real time to provide prompt multi-messenger alerts to other observatories. Aside from enabling prompt SN pointing that is also precise, this will cut down the rate of fake SN triggers (curretnly estimated at ~1/month) and therefore offer potential savings from a reduction in the hardware resources required for storing the large amounts of data associated with CCSN candidates.

System outputs

The output of the AI system is a set of predictions which will be used as the basis for a drastic reduction in the amount of data to be fed to the next stage involving reconstruction and analysis. Before feeding the data to this stage, the AI system will also perform preprocessing operations such as noise removal to facilitate and speed up subsequent data processing.

Technology

AI classification

Classical / predictive MLSource: Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relatio…

System name(s)

Not reported

Custom-developed code

Yes

Public source code

Not reported

Sourcing

How it was built

Not reported

Vendor (as reported)

Not reported

Vendors (standardized)

None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII

No

Privacy Impact Assessment

Not reported

Authorization to Operate

No

Demographic features

Not reported

Training and evaluation data

Simulated data closely approximating real-world raw detector data expected from CCSNs is used to train and validate the ML models used in the data reduction and preprocessing pipeline.

Federal Data Catalog

Not reported

Governance

0 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Potential impacts and how they were identified

Not reported