Andrea Tacchetti is a Senior Research Scientist at DeepMind. His research focuses on Mechanism Design and Multiagent systems. Before joining DeepMind in 2017 Andrea obtained his PhD in Electrical Engineering and Computer Science from MIT.
Arash Mehrjou is a PhD candidate in Max Planck Institute for Intelligent Systems (Tübingen) and ETH Zürich. His interests lie at the intersection of machine learning and dynamical systems and the bidirectional benefits that these disciplines can offer to each other. He has also worked on a number of topics in causality such as causal system identification in reinforcement learning, instrumental variables in counterfactual inference and identifying independent causal mechanisms in generative models.
Ian Gemp is a Senior Research Scientist on the Multiagent team at DeepMind. His research focuses primarily on two questions. How should agents behave in a group, be it a competitive, mixed-motive, or cooperative setting? And should individual agents themselves (including their constituent tools and algorithms) be considered multiagent systems in their own right? He studied mechanical engineering and applied math (BS/MS) at Northwestern University (2011) and obtained his MS/PhD in computer science from the University of Massachusetts at Amherst (2018).
Elise van der Pol
Elise van der Pol is a PhD student in the Amsterdam Machine Learning Lab, working with Max Welling, Frans Oliehoek and Herke van Hoof. Her research interests lie in structure, symmetry, and equivariance in multi-agent and single agent reinforcement learning and machine learning. Elise was an invited speaker at the self-supervision for reinforcement learning workshop at ICLR 2021 and co-organizer of the workshop on ecological/data-centric reinforcement learning at NeurIPS 2021. Before her PhD, she studied Artificial Intelligence at the University of Amsterdam, graduating on the topic of coordination in deep reinforcement learning. She is also involved in UvA's Inclusive AI.
Satpreet H. Singh is a PhD candidate in the Department of Electrical and Computer Engineering at the University of Washington (Seattle) working at the intersection of AI and Computational Neuroscience. He is interested in using theoretical and data-driven approaches to make connections between algorithms used to design intelligent artificial agents and intelligent behaviors seen in biology at different scales (neuron, network, organism, ecology).
Noah Golowich is a PhD student at MIT working on theoretical machine learning. His research aims to address the question of when one can expect self-interested agents interacting in an environment to equilibrate, as well as to understand how the choice of algorithm employed by the agents can improve the speed of convergence to equilibrium. He has also worked on several topics in differential privacy, including the connection between online learnability and private learnability, as well as the shuffled model of differential privacy.
Sarah Perrin a PhD student at Inria Lille, under the supervision of Pr. Olivier Pietquin (Google Brain) and Pr. Romuald Élie (DeepMind). Before starting her PhD, she graduated from École Polytechnique. Her research focuses on applying Reinforcement Learning to solve Mean Field Games. Mean Field Games theory studies decision making in a population of an infinite number of identical agents. It has inspired numerous applications such as population dynamics modeling, crowd motion, economics or energy management and production.
Nina Vesseron is a student in Paris (France) and is currently taking the MVA master's degree, which is a master's degree in applied mathematics in the fields of vision and learning. During her last internships, she worked on how to model neural networks by congestion games as well as on the use of flow-based networks in inverse problem solving. Last year, she obtained her master's degree in Theoretical Computer Science at ENS Lyon (France). She is preparing to do a thesis in AI next year.