Mechanism design with partial revelation

With the emergence of the Internet as a global structure for communication and interaction,
\nmany “business to consumer” and “business to business” applications have migrated online,
\nthus increasing the need for software agents that can act on behalf of people, institutions or
\ncompanies with private and often conflicting interests. The design of such agents, and the
\nprotocols (i.e., mechanisms) through which they interact, has therefore naturally become an 
\nimportant research theme. 
\n
\nClassical mechanism design techniques from the economics literature do not account for the costs 
\nentailed with the full revelation of preferences that they require. The aim of this thesis is to 
\ninvestigate how to design mechanisms that only require the revelation 
\nof partial preference information and are applicable in any mechanism design context. We call this 
\npartial revelation mechanism design. Reducing revelation
\ncosts is thus our main concern. With only partial revelation, the designer has some remaining
\nuncertainty over the agents’ types, even after the mechanism has been executed. Thus, in
\ngeneral, the outcome chosen will not be optimal with respect to the designer’s objective function.
\nThis alone raises interesting questions about which (part of the) information should be
\nelicited in order to minimize the degree of sub-optimality incurred by the mechanism. But this
\nsub-optimality of the mechanism’s outcome choice function has additional important consequences:
\nmost of the results in classical mechanism design which guarantee that agents will
\nreveal their type truthfully to the mechanism rely on the fact that the optimal outcome is chosen.
\nWe must therefore also investigate if, and how, appropriate incentives can be maintained
\nin partial revelation mechanisms.
\n
\nWe start by presenting our own model for partial revelation mechanism design. Our second
\ncontribution is a negative one regarding the quasi-impossibility of
\nimplementing partial revelation mechanisms with exact incentive properties. The rest of the
\nthesis shows, in different settings, how this negative result can be bypassed in various settings, 
\ndepending on the designer's objective (e.g., social welfare, revenue...) and the interaction type 
\n(sequential or one shot). Finally, we study how the approximation of the
\nincentive properties can be further improved when necessary, and in the process, introduce
\nand proves the existence of a new equilibrium concept.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.