September 2, 2021

Announcing the winners of the 2021 RFP on agent-based user interaction simulation to find and fix integrity and privacy issues

By: Meta Research

In May 2021, Facebook launched a request for proposals (RFP) on agent-based user interaction simulation to find and fix integrity and privacy issues. Today, we’re announcing the winners of this award.

View RFP

In this call, we were particularly interested in research work that tackles the scientific and technical challenges we face in developing and deploying Web-Enabled Simulation, and that could also lead to longer and deeper collaboration between partners from the academic research community and our Westworld project team.

This RFP is a continuation of the 2020 RFP on agent-based user interaction simulation. Through these RFPs, Facebook aims to foster further innovation and deepen their collaboration with academia in the area of web-enabled simulation. This year, we narrowed the focus of the technical areas and were interested in proposal topics including, but not limited to, the following:

  • Testing validation and verification
  • Scalability: Cyber-cyber digital twin scalability, prediction, and optimization
  • Behavior: Modeling user behaviors and counterfactual interventions

The team reviewed 29 high-quality proposals from 24 universities in 11 countries. We are pleased to announce the five winning proposals below, as well as the nine finalists. Thank you to everyone who took the time to submit a proposal, and congratulations to the winners.

Research award winners

Automated generation of metamorphic relations for testing WESs
Valerio Terragni, Aitor Arrieta, Paolo Tonella (University of Auckland)

Learning from mistakes to enhance behavioral modeling
Federica Sarro (University College London)

People and models of people
Robert John Nicholls, Monica Whitty, Stephen Doherty (University of New South Wales)

Socialz — Multi-objective automated social fuzz testing
Markus Wagner, Christoph Treude (University of Adelaide)

Testing non-testable programs using association rules
Breno Alexandro Ferreira de Miranda, Antonia Bertolino, Emilio Cruciani, Roberto Verdecchia (Federal University of Pernambuco)


Attacking bad bots via Shapley value approximation and machine learning
Giorgio Stefano Gnecco, Ennio Bilancini, Marcello Sanguineti, Massimo Riccaboni (Scuola IMT Alti Studi Lucca)

Detecting privacy issues in WESs using differential information gain
Partha Roop, Priyadarsi Nanda, Valerio Terragni (University of Auckland)

Evolutionary Bayesian transfer optimization for automated mechanism design
Ke Li (University of Exeter)

Fidelity-aware multi-objective tuner on cyber-cyber digital twin
Tao Chen, Qinggang Meng (University of Loughborough, UK)

Modeling and predicting user behaviors in multi-agent emergent systems
Denys Poshyvanyk, Adwait Nadkarni (College of William and Mary)

Modeling social media platform vulnerabilities
John Bryden, Filippo Menczer (Indiana University Bloomington)

Refining state equivalence relations for effective metamorphic testing
Mike Papadakis, Marcelo d’Amorim, Nazareno Matías Aguirre, Yves Le Traon (University of Luxembourg)

SANS-T: Strategic agents network for social testing
Rocco Oliveto, Simone Scalabrino (University of Molise)

SOCIETY: Social testing by gamified behavior trees
Patrizio Pelliccione, Antonia Bertolino, Michele Flammini (Gran Sasso Science Institute)