Dynamic Resource Allocations without Monetary Transfers

Abstract

In this proposal, following a general research call for proposal on how to design mechanisms that enable efficient sharing of scientific equipment among users, we will investigate settings and conditions under which dynamic mechanism designs without monetary transfers can lead to optimal or near optimal allocation of public resources over time among a set of strategic agents. In particular we will concentrate our attention around three main themes. In the first theme we will consider repeated resource allocation with distributional information about the utilities of the agents. Although a common assumption in the existing literature used to design an allocation mechanism without monetary transfers, we plan to fully investigate the various conditions under which this may lead to optimal designs. In the second theme, we will consider the case of repeated resource allocation without distributional knowledge about the utilities of the agent but with online feedback. Our aim is to understand the interplay between mechanism design and the learning problem of inferring utilities’ distribution. In the third theme, extending the two former research themes, we will consider contextual repeated resource allocation schemes which incorporates some distributional knowledge on agents’ utilities together with online feedback component in a more realistic way.

Document Details

Document Type
DoD Grant Award
Publication Date
Feb 29, 2024
Source ID
FA95502310182

Entities

People

  • Patrick Jaillet

Organizations

  • Air Force Office of Scientific Research
  • Massachusetts Institute of Technology
  • United States Air Force

Tags

Fields of Study

  • Economics

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Operations Research
  • Theoretical Analysis.

Technology Areas

  • AI & ML