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:20201025T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RDATE:20211031T030000
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20210328T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.23351.field_data.0@www.open.diag.uniroma1.it
DTSTAMP:20260407T230943Z
CREATED:20210515T071955Z
DESCRIPTION:Dottorato di Ricerca in Data Science  Neural Information Retrie
 val and NLP Prof. Fabrizio Silvestri (Sapienza Università di Roma) and Pro
 f. Nicola Tonellotto (University of Pisa) Join Zoom Meeting https://unirom
 a1.zoom.us/j/87279867964?pwd=dWpRMk5yRnBqRFVTRzRPd2l3a3BiQT09 Meeting ID: 
 872 7986 7964 Passcode: 433023 20/05 Lesson 1 - Prof. Fabrizio Silvestri f
 rom 9.30 to 13.30 -  Intro to PyTorch\, Language Models\, Implementing Wor
 d2Vec in PyTorchfrom 15.30 to 17.30 - Practicum  21/05 Lesson 2 - Prof. Fa
 brizio Silvestri  from 10.30 to 13.30 - Self-attention\, Transformers\, BE
 RT\, and Beyond. HuggingFace Transformersfrom 15.30 to 17.30 - Practicum  
 27/05 Lesson 3 - Prof. Nicola Tonellotto (Univ. of Pisa) from 9.30 to 13.3
 0 -  Intro to Information Retrieval. Classical models and limitations. Neu
 ral Models for IRfrom 15.30 to 17.30 - Practicum  28/05 Lesson 4 - Prof. N
 icola Tonellotto (Univ. of Pisa)from 9.30 to 13.30 - Neural models for IR1
 5.30 to 17.30 - Practicum  Abstract:Advances from the natural language pro
 cessing community have recently sparked a renaissance in the task of ad-ho
 c search. Particularly\, large contextualized language modeling techniques
 \, such as BERT\, have equipped ranking models with a far deeper understan
 ding of language than the capabilities of previous bag-of-words (BoW) mode
 ls. Applying these techniques to a new task is tricky\, requiring knowledg
 e of deep learning frameworks\, and significant scripting and data munging
 . In this course\, we provide background on classical (e.g.\, BoW)\, moder
 n (e.g.\, Learning to Rank). We introduce students to the Transformer arch
 itecture also showing how they are used in foundational aspects of modern 
 large language models (e.g.\, BERT) and contemporary (e.g.\, doc2query) se
 arch ranking and re-ranking techniques. Going further\, we detail and demo
 nstrate how these can be easily experimentally applied to new search tasks
  in a new declarative style of conducting experiments exemplified by the P
 yTerrier and OpenNIR search toolkits. 
DTSTART;TZID=Europe/Paris:20210515T091500
DTEND;TZID=Europe/Paris:20210515T091500
LAST-MODIFIED:20210515T072601Z
LOCATION:Zoom Meeting https://uniroma1.zoom.us/j/87279867964?pwd=dWpRMk5yRn
 BqRFVTRzRPd2l3a3BiQT09
SUMMARY:Data Science PhD course Neural Information Retrieval and NLP - Prof
 f. Fabrizio Silvestri e Nicola Tonellotto - Prof. Fabrizio Silvestri (Sapi
 enza Univ. of Rome)  and Prof. Nicola Tonellotto (Univ. of Pisa)\n\n\n  \n
   \n\n    \n\n\nFabrizio\n\n\nSilvestri  \n\n  \n\n    \n\n\n\n\n\nProfess
 ore ordinario\n\n\npagina personale\n\nstanza: \n\nB209\n\ntelefono: \n\n+
 39 0677274015\n\nMember of: \n\n  \n\n  \n\n    \n\nBiografia: \n\n\n\nFab
 rizio Silvestri is a Full Professor and the coordinator of the Ph.D. in Da
 ta Science\, at Dipartimento di Ingegneria informatica\, automatica e gest
 ionale (DIAG) of the University of Rome\, La Sapienza. His research intere
 sts lie in Artificial Intelligence\, and in particular\, Fabrizio Silvestr
 i deals with machine learning applied to web search problems and natural l
 anguage processing. He is the author of more than 150 papers in internatio
 nal journals and conference proceedings. It holds nine industrial patents.
  He is the holder of the 'test-of-time' award at the ECIR 2018 conference 
 for an article published in 2007. He is the holder of three best paper awa
 rds and other international awards. Fabrizio Silvestri spent eight years a
 broad in industrial research laboratories (Yahoo! and Facebook). At Facebo
 ok AI\, Fabrizio Silvestri has directed research groups to develop artific
 ial intelligence techniques to combat malicious actors who use the Faceboo
 k platform for malicious purposes (hate speech\, misinformation\, terroris
 m\, etc.) Fabrizio Silvestri has a Ph.D. in computer science awarded by th
 e University of Pisa with a thesis entitled: 'High-Performance Issues in W
 eb Search Engines: Algorithms and Techniques.'\n\n\nqualifica_rr: \n\nProf
 essors
URL;TYPE=URI:http://www.open.diag.uniroma1.it/node/23351
END:VEVENT
END:VCALENDAR