Despite much progress in training artificial intelligence (AI) systems to imitate human
language, building agents that use language to communicate intentionally with humans in …

Blackwell approachability is a framework for reasoning about repeated games with vector-valued
payoffs. We introduce predictive Blackwell approachability, where an estimate of the …

Most existing results about last-iterate convergence of learning dynamics are limited to two-player
zero-sum games, and only apply under rigid assumptions about what dynamics the …

Recently, Daskalakis, Fishelson, and Golowich (DFG) (NeurIPS ‘21) showed that if all agents
in a multi-player general-sum normal-form game employ Optimistic Multiplicative Weights …

We consider the task of accurately modeling strong human policies in multi-agent decision-making
problems, given examples of human behavior. Imitation learning is effective at …

The existence of simple, uncoupled no-regret dynamics that converge to correlated
equilibria in normal-form games is a celebrated result in the theory of multi-agent systems. …

Background Genomewide association studies of autoimmune diseases have mapped hundreds
of susceptibility regions in the genome. However, only for a few association signals has …

The COVID-19 pandemic forced healthcare services organization to adjust to mutating
healthcare needs. Not exhaustive data are available on the consequences of this on non-COVID-…

A genome-wide association scan of ∼6.6 million genotyped or imputed variants in 882
Sardinian individuals with multiple sclerosis (cases) and 872 controls suggested association of …

Background Although IgG oligoclonal bands (OCBs) in the cerebrospinal fluid (CSF) are a
frequent phenomenon in multiple sclerosis (MS) patients, their relationship with grey matter …