Allocation Made PracticaL

Informations projet

AMPLify

N° de convention de subvention: 670077

DOI

10.3030/670077

Projet clôturé le 15 Février 2020

Date de signature de la CE 1 Septembre 2015

Date de début 1 Juin 2016

Date de fin 31 Mai 2021

Financé au titre de

• EXCELLENT SCIENCE - European Research Council (ERC)

Coût total

€ 2 499 681,00

Contribution de l’UE

€ 2 499 681,00

Coordonné par

TECHNISCHE UNIVERSITAT BERLIN
Germany

Periodic Reporting for period 3 - AMPLify (Allocation Made PracticaL)

Période du rapport: 2019-06-01 au 2020-11-30

Résumé du contexte et des objectifs généraux du projet

The AMPLify project tackles a pressing problem facing society today: the efficient and fair allocation of resources and costs. Whilst the foundations of the field of resource and cost allocation have been laid with simple, abstract models, it is now timely to consider richer models that capture features like lack of synchronicity which reflect the practice of resource and cost allocation problems in the real world.The models we are developing are applicable to many new markets being developed in our highly connected and online world. The main goal of the project is to design new mechanisms for resource and cost allocation that exploit the features of these rich new models. The project lies at an exciting interface between three fields: social choice, game theory and optimisation. The emphasis of the project is on computational issues. We are using computation to tackle the complexity of such models and mechanisms, and to increase both efficiency and fairness of allocations. This is of benefit to society as we are able to do more with fewer resources and greater fairness.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

The work performed so far has focused on three specific models of real world resource allocation. All three models share an important (and till now neglected) common feature in that they are dynamic online models in which agents and or resources being allocated are not always present but arrive (and sometimes depart) over time.

The first model is derived from a real world application found around the world where a food bank needs to allocate food as it is donated to the different charities feeding the poor. In this case, the food, the resource being allocated, arrives over time. The second model is derived from the problem of organ matching. When kidneys are donated for transplant, they must be allocated to patients immediately. In this case, both the patients receiving organs, and the donated organs arrive and depart. The third model now under development is derived from the problem of electric car charging. In this case, the cars being allocated charging slots arrive over time.

Progress has been made on the axiomatic level (understanding fundamental axiomatic properties of these three matching markets), on the algorithmic level (designing and studying online mechanisms for these markets), and on the computational level (identifying the complexity of finding optimal and approximate allocations).

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

The rest of the project will be devoted to studying other more realistic allocation settings, and to extracting general lessons that can be applied to other novel and as yet unseen settings.

screen-shot-2019-08-06-at-6-16-40-pm.png