Source author record

Elias Fernández Domingos

Elias Fernández Domingos appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Computer Science and Game Theory cs.CY physics.soc-ph

Catalog footprint

What is connected

2works

3topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2022arXiv

The art of compensation: how hybrid teams solve collective risk dilemmas

It is widely known how the human ability to cooperate has influenced the thriving of our species. However, as we move towards a hybrid human-machine future, it is still unclear how the introduction of AI agents in our social interactions will affect this cooperative capacity. Within the context of the one-shot collective risk dilemma, where enough members of a group must cooperate in order to avoid a collective disaster, we study the evolutionary dynamics of cooperation in a hybrid population made of both adaptive and fixed-behavior agents. Specifically, we show how the first learn to adapt their behavior to compensate for the behavior of the latter. The less the (artificially) fixed agents cooperate, the more the adaptive population is motivated to cooperate, and vice-versa, especially when the risk is higher. By pinpointing how adaptive agents avoid their share of costly cooperation if the fixed-behavior agents implement a cooperative policy, our work hints towards an unbalanced hybrid world. On one hand, this means that introducing cooperative AI agents within our society might unburden human efforts. Nevertheless, it is important to note that costless artificial cooperation might not be realistic, and more than deploying AI systems that carry the cooperative effort, we must focus on mechanisms that nudge shared cooperation among all members in the hybrid system.

preprint2020arXiv

Timing uncertainty in collective risk dilemmas encourages group reciprocation and polarization

Human social dilemmas are often shaped by actions involving uncertain goals and returns that may only be achieved in the future. Climate action, voluntary vaccination and other prospective choices stand as paramount examples of this setting. In this context, as well as in many other social dilemmas, uncertainty may produce non-trivial effects. Whereas uncertainty about collective targets and their impact were shown to negatively affect group coordination and success, no information is available about timing uncertainty, i.e. how uncertainty about when the target needs to be reached affects the outcome as well as the decision-making. Here we show experimentally, through a collective dilemma wherein groups of participants need to avoid a tipping point under the risk of collective loss, that timing uncertainty prompts not only early generosity but also polarized contributions, in which participants' total contributions are distributed more unfairly than when there is no uncertainty. Analyzing participant behavior reveals, under uncertainty, an increase in reciprocal strategies wherein contributions are conditional on the previous donations of the other participants, a group analogue of the well-known Tit-for-Tat strategy. Although large timing uncertainty appears to reduce collective success, groups that successfully collect the required amount show strong reciprocal coordination. This conclusion is supported by a game theoretic model examining the dominance of behaviors in case of timing uncertainty. In general, timing uncertainty casts a shadow on the future that leads participants to respond early, encouraging reciprocal behaviors, and unequal contributions.