BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.adass.org//adass2026//speaker//JQPUAM
BEGIN:VTIMEZONE
TZID:AWST
BEGIN:STANDARD
DTSTART:20000101T000000
RRULE:FREQ=YEARLY;BYMONTH=1;UNTIL=20051231T160000Z
TZNAME:AWST
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
END:STANDARD
BEGIN:STANDARD
DTSTART:20070325T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:AWST
TZOFFSETFROM:+0900
TZOFFSETTO:+0800
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20071028T030000
RRULE:FREQ=YEARLY;BYDAY=4SU;BYMONTH=10;UNTIL=20081025T190000Z
TZNAME:AWDT
TZOFFSETFROM:+0800
TZOFFSETTO:+0900
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-adass2026-MGXQW3@pretalx.adass.org
DTSTART;TZID=AWST:20261102T102700
DTEND;TZID=AWST:20261102T103600
DESCRIPTION:The New Space Economy\, enabled by the increasing availability 
 of commercial launch vehicles and spacecraft\, has renewed interest in lun
 ar exploration. Within the Italian PRORIS program and the HARLOCK project\
 , the Rare-earth & Heavy Elements Analyzer (RHEA) is being developed. RHEA
  is a payload consisting of a suite of instruments designed for integratio
 n on small rovers\, enabling rare earth element (REE) and rare-metal scout
 ing and prospecting through a highly coordinated and synergistic approach 
 based on a progressive and multiscale observation strategy. RHEA combines 
 gamma-ray spectroscopy\, hyperspectral imaging\, ultraviolet fluorescence 
 screening\, and X-ray fluorescence analysis to identify and characterize R
 EE-bearing deposits and associated elemental and mineralogical signatures 
 across different spatial scales. Its operation in the lunar environment is
  constrained by limited onboard power and communication bandwidth with Ear
 th\, making artificial intelligence (AI) a key enabling technology for aut
 onomous scientific operations and onboard data processing.\n\nWe present t
 he AI-based Real-time Computational Analysis platform for Decision Inferen
 ce in Autonomous space systems (ARCADIA)\, which implements the Instrument
  Control Unit (ICU) and onboard Data Processing Unit (DPU) for the RHEA ro
 ver. The ICU controls and monitors the scientific payload\, while the DPU 
 performs AI-enhanced onboard data analysis\, including spectral identifica
 tion of REEs\, real-time multi-instrument data fusion\, and data reduction
 . These capabilities support the generation of scientific products\, auton
 omous decision-making\, and efficient use of communication bandwidth and o
 nboard storage. Real-time onboard processing of scientific data enables AR
 CADIA to enhance the effectiveness of the scouting process by rapidly guid
 ing the rover toward the most promising sites for further investigation\, 
 thereby accelerating target selection and exploration even during short-du
 ration missions.\n\nThe ARCADIA software architecture includes interfaces 
 with the rover and scientific payloads. A market survey of low-power Comme
 rcial Off-The-Shelf (COTS) computing platforms was conducted to identify s
 olutions meeting the project requirements for power consumption\, computat
 ional capability\, and environmental constraints. AI algorithms for onboar
 d data analysis were evaluated on low-power edge-computing platforms\, inc
 luding the NVIDIA Jetson Nano\, and on RISC-V processors representative of
  flight-proven hardware. Candidate platforms target a power consumption be
 low 5 W while supporting AI models optimized through pruning and quantizat
 ion\, with the goal of enabling real-time onboard processing under lunar r
 esource constraints.\n\nThe technologies and methodologies developed withi
 n ARCADIA are designed to be transferable to other autonomous space system
 s and adaptable to a broader range of future space missions.
DTSTAMP:20261116T153711Z
LOCATION:Seminar Room 2
SUMMARY:ARCADIA: AI-based Real-time Analysis for Autonomous Space Systems -
  Nicolò Parmiggiani\, Ghulam Mustafa
URL:https://pretalx.adass.org/adass2026/talk/MGXQW3/
END:VEVENT
BEGIN:VEVENT
UID:pretalx-adass2026-WWKV97@pretalx.adass.org
DTSTART;TZID=AWST:20261102T153600
DTEND;TZID=AWST:20261102T154800
DESCRIPTION:In astrophysics\, Deep neural networks have achieved state-of-t
 he-art performance in a variety of spectral analysis tasks. However\, thes
 e models are often considered "black boxes"\, providing limited insight in
 to predictions and requiring post-hoc tools for explanations. This lack of
  transparency can limit their adoption in scientific contexts\, as in thes
 e environments interpretability is essential. This is one of the challenge
 s in the development of RHEA\, the Rare-earth \\& Heavy Elements Analyzer 
 of the HARLOCK project within the PRORIS program. It combines gamma-ray sp
 ectroscopy\, multispectral imaging\, ultraviolet fluorescence screening an
 d X-ray fluorescence analysis to identify Rare-Earth Elements (REE) deposi
 ts and associated elemental and mineralogical signatures across different 
 spatial scales. \n\nIn this work\, we propose a model-agnostic framework t
 hat converts any Convolutional spectral classifier into an explainable pre
 dictor without hindering its performance on multiple metrics\, demonstrati
 ng it on a complete model working with attention masks for additional insi
 ght and training stability. Our approach exploits the Neural Network's lat
 ent representation and preservation of spatial features: after training\, 
 the final regression layer of the backbone model is removed and replaced b
 y a K-Nearest Neighbors (KNN) regressor operating directly on the learned 
 embedding space\; as a result\, every prediction is accompanied by an expl
 anation in the form of the most similar spectra observed during training. 
 This works particularly well thanks to a custom loss function designed to 
 optimize the quality of the learned embedding space\, without hindering pe
 rformances. \n\nFurthermore\, in our model\, for each predicted element an
  associated attention mask is learned at training time starting from a phy
 sics-based initialization. This way\, the framework provides\, alongside i
 nstance-level explanations\, easily readable heatmaps in the form of a mas
 k-by-weights product\, allowing users to trace model outputs back to repre
 sentative examples in the training data and the relevant parts of the inpu
 t used. Experimental results show that\, in a regression task estimating t
 he composition of soil samples\, the average R2 score over the predicted e
 lements stays around 0.9 for both the original approach and the explainabl
 e Deep kNN model (depending on the complexity of the task).\n\nOur proposa
 l offers a practical path toward trustworthy AI for spectroscopy\, combini
 ng the representational power of deep learning with the transparency of ex
 ample-based reasoning.
DTSTAMP:20261116T153711Z
LOCATION:Seminar Room 2
SUMMARY:Explainable XRF Element Detection with Embedding-Space KNN and Phys
 ics-Based Masks for the HARLOCK Lunar Rover - Alberto Luise\, Ghulam Musta
 fa
URL:https://pretalx.adass.org/adass2026/talk/WWKV97/
END:VEVENT
BEGIN:VEVENT
UID:pretalx-adass2026-KR9DUR@pretalx.adass.org
DTSTART;TZID=AWST:20261102T154800
DTEND;TZID=AWST:20261102T160000
DESCRIPTION:Terrestrial Gamma-ray Flashes (TGFs) are sub-millisecond bursts
  of atmospheric gamma rays associated with thunderstorms and are of partic
 ular interest for space-based gamma-ray observations. Their detection from
  orbit requires the identification of brief gamma-ray signals in the prese
 nce of background\, while operating under the limited computational and po
 wer resources typically available on CubeSat-class spacecraft. Within this
  context\, the Gamma-Sky project is investigating low-power Edge-AI approa
 ches for the real-time processing of high-energy radiation\, using atmosph
 eric X- and gamma-ray detector data. This work presents a machine-learning
 -based event-classification methodology for a CubeSat-class mission\, expl
 oiting the spatial and energy information provided by a segmented scintill
 ator detector architecture. The proposed detector consists of multiple sci
 ntillator elements distributed throughout the detector volume rather than 
 a single monolithic crystal. This configuration enables the identification
  of characteristic event patterns related to deposited energy\, multiplici
 ty of interactions\, and spatial correlations between detector elements. T
 he detector response is used to extract a compact set of event-level featu
 res suitable for machine-learning analysis. A dedicated algorithm is then 
 developed to distinguish gamma-ray interactions from charged-particle-indu
 ced background\, reducing the charged-particle contamination of a TGF even
 t. The simulation results show that the spatial and energy-deposition info
 rmation provided by the segmented detector offers effective discrimination
  between gamma-ray interactions and charged-particle-induced events. The p
 rocessing chain is designed for resource-constrained onboard operation\, w
 ith particular attention to reducing the computational and memory requirem
 ents of the trained models. Quantization and pruning are applied to obtain
  a lightweight implementation while preserving the event-classification ca
 pability. The resulting algorithm is suitable for deployment and experimen
 tal validation on embedded Edge-AI platforms. Preliminary results indicate
  the potential of combining detector segmentation\, machine-learning-based
  event classification for background rejection\, and efficient onboard pro
 cessing in future compact high-energy space instruments dedicated to TGF o
 bservations.
DTSTAMP:20261116T153711Z
LOCATION:Seminar Room 2
SUMMARY:Machine-Learning-Based Event Classification of Terrestrial Gamma-ra
 y Flashes with a Segmented Scintillator Detector - Antonio\, Ghulam Mustaf
 a
URL:https://pretalx.adass.org/adass2026/talk/KR9DUR/
END:VEVENT
BEGIN:VEVENT
UID:pretalx-adass2026-N79KYE@pretalx.adass.org
DTSTART;TZID=AWST:20261103T101200
DTEND;TZID=AWST:20261103T102400
DESCRIPTION:The Compton Spectrometer and Imager (COSI) is a NASA Small Expl
 orer mission designed to study the MeV gamma-ray sky\, including gamma-ray
  bursts (GRBs)\, compact objects\, nuclear emission lines\, and other high
 -energy sources. Reliable background prediction is important for transient
  searches because the background changes with spacecraft position\, pointi
 ng direction\, and time. The aim of this paper is to develop a probabilist
 ic machine-learning framework for COSI background prediction to support GR
 B detection and localization. The framework consists of four main stages. 
 The Angular Resolution Measure (ARM) method is used to filter the event da
 ta and generate light curves for different sky positions following the HEA
 LPix grid. These background light curves are then used to train a probabil
 istic regression model and derive significance thresholds from its predict
 ive distribution. The trained model is applied to the same sky positions w
 ithin the COSI field of view to estimate the expected background. The obse
 rved light curves are then compared with the predicted background and sign
 ificance thresholds to search for significant excesses associated with GRB
 s and support their detection and localization. In this work\, we focus on
  developing and evaluating the background regression models using simulate
 d DC4 background events and spacecraft information. For the regression ana
 lysis presented here\, we use an ARM-selected background light curve with 
 one-second bins. Spacecraft livetime information is used to select valid o
 bserving periods and remove bins close to South Atlantic Anomaly boundarie
 s. The model inputs include the time elapsed since the start of the select
 ed DC4 modeling interval\, spacecraft altitude\, the Earth-zenith directio
 n\, and spacecraft pointing orientation. The spacecraft pointing direction
 s are originally given in Galactic coordinates and are represented in the 
 models using Cartesian components. From the DC4 dataset\, two consecutive 
 orbits are reserved as a held-out temporal test block\, while the remainin
 g data are randomly divided into training and test sets. Only background e
 vents are used for model training and evaluation. We compare Poisson regre
 ssion with a four-component Gaussian Mixture Regression (GMR) model\, with
  the GMR providing a more flexible predictive distribution. The current re
 sults show that both models follow the large-scale evolution of the ARM-se
 lected background. As future work\, we plan to evaluate the trained backgr
 ound prediction models using simulated GRBs to measure their performance f
 or transient detection and localization.
DTSTAMP:20261116T153711Z
LOCATION:Seminar Room 2
SUMMARY:Machine Learning Models for Predicting the COSI Background for Gamm
 a-Ray Burst Detection - Ghulam Mustafa
URL:https://pretalx.adass.org/adass2026/talk/N79KYE/
END:VEVENT
BEGIN:VEVENT
UID:pretalx-adass2026-C7GXZ3@pretalx.adass.org
DTSTART;TZID=AWST:20261104T101200
DTEND;TZID=AWST:20261104T102400
DESCRIPTION:The detection and localization of gamma-ray sources is a comple
 x optimization problem\, requiring the estimation of parameters such as th
 e source flux\, background\, position\, and detection significance. AGILE 
 is an Italian Space Agency mission whose Gamma-Ray Imaging Detector (GRID)
  surveyed the sky in the 30 MeV-50 GeV band and produced photon count maps
 . Detecting gamma-ray sources in this data can be challenging because of l
 ow photon counts\, which require specific methods to separate the source f
 rom the diffuse background. For this reason\, the classical approach to de
 tect and localize gamma-ray sources is the Maximum Likelihood Estimation (
 MLE). This fitting problem can be seen as an optimization over the model p
 arameters\, making it a candidate for quantum annealing\, a computational 
 paradigm that solves optimization problems by encoding them into the groun
 d state of an Ising Hamiltonian. This technique does so through a slow\, c
 ontrolled process that evolves into the minimum-energy configuration\, whi
 ch then encodes the solution to the optimization problem\, representing a 
 promising approach in the current Noisy Intermediate-Scale Quantum era. In
  this work\, we reformulate the standard likelihood analysis of AGILE-GRID
  data as a Quadratic Unconstrained Binary Optimization (QUBO) problem\, ma
 king the cost function quadratic and expressing the model parameters in bi
 nary form\, as required by a quantum annealer. This approach could speed u
 p the analysis of complex data\, where the classical MLE fit becomes slow 
 or may fail to converge.\nWe solve this task with simulated quantum anneal
 ing and benchmark it against the classical pipeline implemented in AGILEpy
 \, the official analysis library for the AGILE mission\, on 200 simulated 
 observations at extragalactic latitude. Both the classical and quantum app
 roaches recover the source flux accurately\, with a systematic error of +2
 .8% for the former and -0.3% for the latter. For the isotropic background\
 , the mean raw residuals are -0.87 (QUBO) and -0.58 (classical). We also f
 ound that the QUBO identifies the correct position among the 81 candidate 
 positions on a grid with of 0.5° resolution in 90% of the maps. These res
 ults show that the quantum approach achieves accuracy close to that of the
  classical pipeline\, a promising result for applying quantum annealing to
  gamma-ray source detection and localization. As a next step\, we plan to 
 test the method on real quantum annealing hardware.
DTSTAMP:20261116T153711Z
LOCATION:Seminar Room 2
SUMMARY:Quantum Annealing for Gamma-Ray Source Detection and Localization w
 ith AGILE Data - Alessandro Rizzo\, Ghulam Mustafa
URL:https://pretalx.adass.org/adass2026/talk/C7GXZ3/
END:VEVENT
END:VCALENDAR
