Building a Legacy: LSST, FASTLab, and the Privilege of an Academic Platform.

Associate Professsor

University of Delaware

Department of Physics and Astronomy

Biden School of Public Policy and Administration

Data Science Institute

LSST Survey Scientist

federica b. bianco

she/her

Grad student

Postdoc

UG student

NSF

$4.8M

other

$1.2M

NASA

$400K

DOE

$500K

NIH

$80K

Astro

$3.4M

Data Science

$1.6M

Inter

disciplinary

$1.7M

experiment driven science -∞:1900

theory driven science 1900-1950

data driven science 1990-2010

the fourth paradigm - Jim Gray, 2009

computationally driven science 1950-1990

experiment driven science -∞:1900

theory driven science 1900-1950

data driven science 1990-2010

the fourth paradigm - Jim Gray, 2009

computationally driven science 1950-1990

AI driven science? 2010...

The embryo - the Weather Buerau's $2,000,000 "704" computer - learned to differentiate between left and right after 50 attempts in the Navy demonstration

NEW NAVY DEVICE LEARNS BY DOING; Psychologist Shows Embryo of Computer Designed to Read and Grow Wiser

July 8, 1958

1958

The embryo - the Weather Buerau's $2,000,000 "704" computer - learned to differentiate between left and right after 50 attempts in the Navy demonstration

The Navy revealed the embryo of an electronic computer today that it expects will be able to walk, talk, see, write, reproduce itself and be conscious of its existence.

NEW NAVY DEVICE LEARNS BY DOING; Psychologist Shows Embryo of Computer Designed to Read and Grow Wiser

July 8, 1958

1958

The Nobel Prize in Physics 2024 was awarded jointly to John J. Hopfield and Geoffrey Hinton "for foundational discoveries and inventions that enable machine learning with artificial neural networks"

The Nobel Prize in Physics 2024 was awarded jointly to John J. Hopfield and Geoffrey Hinton "for foundational discoveries and inventions that enable machine learning with artificial neural networks"

There are classes of physical systems whose spontaneous behavior can be used as a form of general (and error-correcting) content-addressable memory.

Consider the time evolution of a physical system that can be described by a set of general coordinates. A point in state space then represents the instantaneous condition of the system. This state space may be either continuous or discrete (as in the case of N Ising spins). The equations of motion of the system describe a flow in state space. ....

1980s...

number of arXiv:astro-ph submissions with abstracts containing one or more of the strings: ‘machine learning’, ‘ML’, ‘artificial intelligence’, ‘AI’, ‘deep learning’ or ‘neural network’.

The claim that so-called artificial intelligence (AI) can gain consciousness is on the verge of becoming mainstream. [...]

We argue that the association between consciousness and the computer algorithms used today [...], is deeply flawed.

We believe that these flawed associations arise from a lack of technical knowledge and the way several new technologies (especially LLMs) work

July 8, 1958

2025

To advance Astrophysical Discoveries (with AI and new instrument systems)

To Understand AI and complex systems

To advance Astrophysical Discoveries (with AI and new instrument systems)

To Understand AI and complex systems

To increase AI literacy (to save the world)

To advance Astrophysical Discoveries (with AI and new instrument systems)

To Understand AI and complex systems

To increase AI literacy (to save the world)

To use technical and collaboration skills for good

LSST:

The Vera C. Rubin Observatory Legacy Survey of Space and Time

20Tb of data every night. That is equivalent to

8,000 high definition movies

4,000 hours of tiktok videos

every night for 10 years

what's in a name?

The first ground-based national US observatory named after a woman, Dr. Vera C. Rubin

what's in a name?

what's in a name?

The first ground-based national US observatory named after a woman, Dr. Vera C. Rubin

“The most important feature of any telescope is the imagination with which it is used.”

1996-1998 Tony Tyson, Roger Angel

How it started

with Zhoran Mandami,

Astronaut Reid Wiseman,

Activist Zabib Musa Loro,

Pope Leo XIV,

Olympic medalist Alysa Liu,

Benicio Del Toro.......

January 2016

March 2017

@fedhere

Fabruary 2020

@fedhere

May 2022

June 2023

The DOE LSST Camera - 3.2 Gigapixel

The LSST 20-year Observing Strategy: a sociotechnical problem

6 filters, 18,000 sq deg, 800 repeat observations, 1800 pointings, 5 fields with 8000 pointings, >3 special programs, 10 years

The Astrophysical Journal Supplement Special Issue on LSST Survey Strategy

PI Bianco

Text

Rubin ToO Publications

MacBride, w Bianco and Acero Cuellar 2026 SPIE
MacBride, w Bianco and Acero Cuellar 2026 GCN

Annan, w Bianco and Acero Cuellar 2026 GCN
MacBride, w Bianco and Acero Cuellar 2026 GCN
Andreoni, w Bianco, The Astronomical Journal Supplements, 2026 (in review)

Rubin ToO Program for GW follow up

Eye to the sky…on-sky engineering tests have begun at @nsfgov–@energy Rubin Observatory using the world’s largest digital camera!🔭

April 17 2025

Eye to the sky…on-sky engineering tests have begun at

Rubin Observatory using the world’s largest digital camera!

June 23 2025

First Look party here at UD with 213 people signed up!

Organized by FASTLab

678 separate images taken in just over seven hours of observing time. Trifid nebula (top right) and the Lagoon nebula, which are several thousand light-years away from Earth. | NSF-DOE Vera C. Rubin Observatory

Acero-Cuellar+

The Astronomical Journal, 2026

Volume 171, Issue 6, id.360, 43 pp.

Acero-Cuellar et al. DESC submitted

Tatiana Acero-Cuellar

UNIDEL fellow

LSST Data Science Fellow, UD DSI Fellow

rapid accutare alert release

in 60 seconds:

Difference Image Analysis

template

in 60 seconds:

Difference Image Analysis

template

difference image

in 60 seconds:

Difference Image Analysis

template

difference image

Rubin Observatory Data Management Team

federica bianco - fbianco@udel.edu

Can a NN replace difference imaging?

What is the AI learning and what can we learn from the AI?

template

difference

template

template

difference

template

Can a NN replace difference imaging?

What is the AI learning and what can we learn from the AI?

Saliency maps: what pixels matter?

template

difference

The Rubin LSST ML-Reliability Score (aka real-bogus)

Acero-Cuellar+ in prep 2026

Weston, Acero-Cuellar+ in prep 2026

Bayesian analysis of citizens' labels

https://www.zooniverse.org/projects/ebellm/rubin-difference-detectives

Mathilda Nilsson

second year

DPA Ph.D. candidate

exploring evolutionary tracks of DP2 detections in a space of familiar transients

what can we find?

Easton Honaker

NASA Space Grant Fellow

from John's group

Dr. Riley Clarke, DPA graduate
Now Postdoc at the University of Singapore

LSST Discovery Alliance Grant

through the DSI ($17,000)

written 100% by the FASTLab students!

Text

The Astrophysical Journal Letters, 2026

Volume 994, Issue 1, id.L24, 9 pp.

Why do we study stellar explosions?

we are made of stars

The nitrogen in our DNA, the calcium in our teeth, the iron in our blood, the carbon in our apple pies were made in the interiors of collapsing stars.

We are made of starstuff.

― Carl Sagan, Cosmos

Farthest: 10.5 billion years ago

3 billion years after the Big Bang

redshift 4

Why do we study stellar explosions?

they are the best tool to "measure" the Universe

largest explosion on earth 10,000,000 erg

typical supernova....

Why do we study stellar explosions?

a unique opportunity to study extreme energy events

who'sploded?

10 stars explode in the universe every second

Until the 1900s we would see 1 in a century

Until the 1980s we would see 1 in a decade

Until the 2010s we would see 1 in a month

With the Vera C. Rubin Observatory we will see 1000 every night !

Rubin LSST Transients by the numbers

17B stars (x10) Ivezic+19

~10 million QSO (x10) Mary Loli+21

~50k Tidal Disruption Events (from ~150) Brickman+ 2020

~10k SuperLuminous Supernovae (from ~200) Villar+ 2018

~400 strongly lensed SN Ia (from 10) Ardense+24

~50 kilonovae (from 2) Setzer+19, Andreoni+19 (+ ToO)

> 10 Interstellar Objects fom 2.... ?)

SKA

(2025)

17B stars (x10) Ivezic+19

~10 million QSO (x10) Mary Loli+21

~50k Tidal Disruption Events (from ~150) Brickman+ 2020

~10k SuperLuminous Supernovae (from ~200) Villar+ 2018

~400 strongly lensed SN Ia (from 10) Ardense+24

~50 kilonovae (from 2) Setzer+19, Andreoni+19 (+ ToO)

> 10 Interstellar Objects fom 2.... ?)

True Novelties!

Rubin LSST Transients by the numbers

"BUT BIG DATA DOES NOT MEAN BIG SCIENCE"

Yang Huang,
University of Chinese Academy of Sciences

SpecCLIP talk @UNIVERSAI

IAU workshop Greece June 2025

Challange

follow up

federica bianco - fbianco@udel.edu

Rubin will see ~1000 SN every night!

Credit: Alex Gagliano IAIFI fellow MIT/CfA

- There are 100s of 1-2 m class telescopes
mag limit ~19-22; %1 of the LSST events

- There are ~20 2-4m class telescopes

mag limit 22-24; %10 of the LSST events

- There are ~6 8m class telescopes

mag limit 24+; %100 of the LSST events

1000 SN in the LSST sky every night

Text

The first systematic study of resolution and signal requirements for spectral classification of supernovae

under review, The Open Journal of Astrophysics

arxiv.org/pdf/2607.03532a

Dr. Willow Fox Fortino

A new AI-based classifier for SN spectra at low resolution based on generative AI and transformers

Dr. Willow Fox Fortino

Ally Baldelli,

second year

DPA Ph.D. candidate

Poster presented ar:

Rubin Community Workshop, 2026

Time Domain Astronomy with LSST, 2026

new data,

more transient classes,

state of the art (hierarchical) loss function

Paula Gálvez Molina

DPA student with Frank

now in Dissertation Status!!

What kind of data encoding works for pretrained models?

spectral data visual encoding

Challange

data encoding

so you wanna study galaxies (or 2d nano materials in Lars's lab...)

visualizatoin and concept credit: Alex Razim

so you wanna study galaxies (or 2d nano materials in Lars's lab...)

visualizatoin and concept credit: Alex Razim

Kaicheng Zhang et al 2016 ApJ 820 67

SN 2011fe

Boone 2017

7% of LSST data

Boone 2017

7% of LSST data

The rest

Text

https://arxiv.org/pdf/2405.01672

we introduce

Gaussian process Optimized Photometric Regression of Extragalactic Archival Ultraviolet-infrared eXplosions, a.k.a GOPREAUX—

a Python package for Gaussian Process Regression of multi-wavelength transient photometry. [...]

This allows for predictions of light curves [...] at higher redshifts, where the rest-frame UV emission is redshifted into the observer-frame optical or infrared.

under review, ApJ

Text

Dr. Somayeh Khakpash

LSST Catalyst Fellow

Lehigh University Visiting Prof.

Gaussian processes work by imposing a kernel that represents the covariance in the data (how data depend on time or time/wavelength). Imposing the same kernel for different time-domain phenomena is principally incorrect

=> bias toward known classes

Methodological issues with these approaches

Neural processes replace the imposed kernel with a learned one

Siddharth Chaini

NASA FINESST Fellow

UD Excellence Fellow

DSI Fellow

under review, NEURIps

Siddharth Chaini

NASA FINESST Fellow

UD Excellence Fellow

under review, NEURIps

Siddharth Chaini

NASA FINESST Fellow

UD Excellence Fellow

Can the latent space of Neural Processes be used for represntation?

Adeoluwa Afolabi Augustina Lorda Abdubrahim Silla

Lincoln University Research Interns

Siddharth Chaini

NASA FINESST Fellow

UD Excellence Fellow

Ensambles of distance metrics are excellent for anomaly detection under a constrained budget:

they find true out of class events

The violent and rapidly varying radiation from black holes, neutron stars, and white dwarfs makes them promising targets for high time resolution imaging.

The rotation, pulsation, and local accretion dynamics of these compact stellar remnants tends to occur on timescales ranging from seconds to milliseconds. Their extreme densities also makes them an excellent testing ground for nuclear, quantum, and gravitational physics.

Thomas and Kahn, 2018

Additional targets