Seminario Pubblico di Federico Fusco
NELL’AMBITO DELLA PROCEDURA SELETTIVA DI CHIAMATA, AI SENSI DEL NOVELLATO ART. 24, COMMA 3, LEGGE 240/2010, PER N. 1 POSTO DI RICERCATORE A TEMPO DETERMINATO IN TENURE TRACK (RTT) PER IL GRUPPO SCIENTIFICO DISCIPLINARE 09/IINF-05 - SETTORE SCIENTIFICO DISCIPLINARE IINF-05/A, PRESSO IL DIPARTIMENTO DI INGEGNERIA INFORMATICA, AUTOMATICA E GESTIONALE “ANTONIO RUBERTI” – FACOLTÀ DI INGEGNERIA DELL’INFORMAZIONE, INFORMATICA E STATISTICA
FEDERICO FUSCO TERRA’ UN SEMINARIO PUBBLICO IN DATA 17/3/2025, ALLE ORE 14:00 PRESSO L’AULA B203 DEL DIAG, E IN COLLEGAMENTO Zoom, link alla videochiamata:
https://uniroma1.zoom.us/j/81925745590?pwd=3bpMj5ch1ezLLGaQQeP46zX9ONhbZg.1
Title: Online Learning and Economics
Abstract It is common in economic theory to model the intrinsic uncertainty of reality with random variables whose laws are perfectly and publicly known. However, practical applications often call for a more realistic and data-driven approach, as the relevant features of the problem at hand must be obtained on the field. The online learning framework provides a natural step in this direction, where no initial information is assumed, and data is collected online by repeatedly interacting with the environment. Online learning is particularly suited for economic applications as it captures the exploration-exploitation tradeoff and the intriguing feedback models that may naturally arise. This talk surveys recent works on online learning and economics, focusing on bidding and auction design. The goal is to provide a perspective on the technical challenges faced in designing and analyzing learning algorithms and to offer the tools to address them. Part of the talk is devoted to the construction of “hard instances” as a way of proving tightness results. The talk is partly based on the recent papers “The Role of Transparency in Repeated First-Price Auctions with Unknown Valuations” (STOC 24), “No-Regret Learning in Bilateral Trade via Global Budget Balance” (STOC 24), and “Selling Joint Ads: A Regret Minimization Perspective” (EC 24).
Short Bio Federico Fusco is an RTDA at the Department of Computer, Control, and Management Engineering at Sapienza University of Rome. Previously, he was a PostDoc and completed his PhD under the supervision of Stefano Leonardi in the same institution. During his Ph.D., he was hosted by Paul Duetting at Google Research Zurich as a research intern and then as a student researcher. Federico's research interests span Algorithmic Game Theory, Online Learning, and Submodular Maximization.