BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:Eventi DIAG
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:STANDARD
DTSTART:20261025T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20260329T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RDATE:20270328T020000
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.31453.field_data.0@www.diag.uniroma1.it
DTSTAMP:20260915T152855Z
CREATED:20260907T084740Z
DESCRIPTION:Speaker: Daniele Malitesta (LUISS Guido Carli University)Title:
  Diffusion-Based Multimodal Graph Machine Learning Under Unreliable Graph 
 DataZoom link: https://uniroma1.zoom.us/j/81178556941?pwd=7qp2VjUNNTOtA0A5
 cy95eaQkN9bRGo.1 AbstractGraphs are a natural representation for a wide ra
 nge of real-world data\, from social networks and e-commerce interactions 
 to molecules and AI agent communities. Graph neural networks have become t
 he standard model for learning from such non-Euclidean structures\, and th
 eir extension to multimodal settings (where nodes\, edges\, or features ca
 rry information from images\, text\, or other modalities) has opened new o
 pportunities across domains such as personalized recommendation\, healthca
 re\, and agent-based task planning. Nevertheless\, multimodal graph data c
 ollected in real-world scenarios is rarely clean: it is often incomplete\,
  noisy\, or inconsistent\, undermining the reliability of downstream graph
  learning models. This talk presents my recent and current research on mul
 timodal graph machine learning under this broader challenge of unreliabili
 ty\, and discusses how graph generative models can offer a principled way 
 to address it\, with a particular focus on the personalized recommendation
  domain. Building on this\, I will outline a future research agenda center
 ed on graph diffusion models (a notable and recent example of powerful gra
 ph generative approaches) and their potential to tackle unreliable multimo
 dal graph data. This agenda spans three directions: measuring unreliabilit
 y in multimodal graph data\; designing diffusion-based strategies to mitig
 ate it\, both as a standalone correction step and as an integrated compone
 nt of downstream pipelines\; and understanding the limitations of diffusio
 n models themselves\, including scalability and fairness concerns. I will 
 conclude by discussing how these directions could generalize beyond recomm
 endation to other domains\, such as healthcare and the emerging area of gr
 aph-based task planning for AI agents (a direction I am currently explorin
 g at LUISS). Zoom meeting: https://uniroma1.zoom.us/j/81178556941?pwd=7qp2
 VjUNNTOtA0A5cy95eaQkN9bRGo.1  BioDaniele Malitesta is a postdoctoral resea
 rcher at LUISS Guido Carli University. Previously\, he held a postdoctoral
  position at Université Paris-Saclay\, and before that he completed his Ph
 D in Computer Science Engineering at Politecnico di Bari. His research spa
 ns graph machine learning\, personalized recommendation\, and multimodal d
 eep learning\, with a recent focus on multimodal graph learning and genera
 tion under unreliable graph data. His work has been published in leading c
 onferences\, including AISTATS\, The Web Conference\, ACM Multimedia\, SIG
 IR\, and CIKM\, where he has also been nominated outstanding reviewer thre
 e times\, as well as in high-impact journals such as IEEE Transactions on 
 Knowledge and Data Engineering and Expert Systems With Applications. Danie
 le is active in the research community\, organizing venues such as the Lea
 rning on Graphs Conference and the DaQuaMRec workshop at ACM RecSys\, and 
 giving invited talks\, tutorials\, and summer school lectures. He recently
  joined the editorial board of the journal Intelligenza Artificiale and cu
 rrently serves as guest editor for two Special Issues\, at Data Mining and
  Knowledge Discovery and ACM Transactions on Recommender Systems.
DTSTART;TZID=Europe/Paris:20260911T140000
DTEND;TZID=Europe/Paris:20260911T140000
LAST-MODIFIED:20260907T091028Z
LOCATION:Zoom link: https://uniroma1.zoom.us/j/81178556941?pwd=7qp2VjUNNTOt
 A0A5cy95eaQkN9bRGo.1
SUMMARY:Seminar Daniele Malitesta - September 11\, 14:00 - Online - Daniele
  Malitesta
URL;TYPE=URI:https://www.diag.uniroma1.it/node/31453
END:VEVENT
END:VCALENDAR