The Centre for Humans and Machines invites interested attendees to our public seminars, which feature scientists from our institute and experts from all over the world. Our seminars usually take 1 hour and provide an opportunity to meet the speaker afterwards.
Delegation to autonomous agents promotes cooperation in collective risk dilemmas
Talk by Elias Fernández Domingos, Vrije Universiteit Brussel
29 Jun, 2021 at 15:30 p.m. (CET)
Meeting number: 163 440 3385
Home assistant chat-bots, self-driving cars, drones or automated negotiations are some of the several examples of autonomous (artificial) agents that have pervaded our society. These agents enable the automation of multiple tasks, saving time and (human) effort. However, their presence in social settings raises the need for a better understanding of their effect on social interactions and how they may be used to enhance cooperation towards the public good, instead of hindering it. To this end, we present an experimental study of human delegation to autonomous agents and hybrid human-agent interactions centred on a non-linear public goods dilemma with uncertain returns in which participants face a collective risk. Our aim is to understand experimentally whether the presence of autonomous agents has a positive or negative impact on social behaviour, fairness and cooperation in such a dilemma. Our results show that cooperation increases when participants delegate their actions to an artificial agent that plays on their behalf. Yet, this positive effect is reduced when humans interact in hybrid human-agent groups. Also, we show that humans have biases towards agent behaviour, assuming that they will contribute less to the collective effort. In general, we find that delegation to autonomous agents might act as long-term commitment devices, which prevent both the temptation to deviate to an alternate (less collectively good) course of action, as well as limiting responses based on betrayal aversion.
Elias Fernández Domingos is a Postdoctoral Researcher at the Machine Learning Group of the ULB (Belgium) and a Research Fellow at the Artificial Intelligence Laboratory of the VUB (Belgium). Elias completed his doctoral studies at the VUB. He has also done research stays at the AtlanTic research group at the University of Vigo (Spain) and at the Group for People and Society part of the INESC-ID (Portugal). His research combines game theory, evolutionary dynamics and machine learning to study and model social dynamics in large populations. Alongside theoretical and computational models, he has performed distinct behavioral economics experiments with more than 400 participants.
An Online Game Platform to Study Systemic Sustainability of Common Pool Resources with Environmental Feedback
Talk by Samir Simon Suweis, University of Padova
06 Jul, 2021 at 15:30 p.m. (CET)
Meeting number: 163 883 9899
The sustainable use of common-pool resources (CPRs) is a major environmental governance challenge because of their risk of over-exploitation due to short-term profit-maximization. Communities may devise self-governing institutions that avoid overuse and attain the long-term benefits of cooperation. It is still unclear what conditions allow cooperation to emerge, leading to greater long-term rewards. Until recently, the study of the sustainable governance of common pool resources has overlooked the feedback between user decisions and resource dynamics, while the ability of shared goals to induce cooperation still needs to be tested. Here we develop an online game to simulate a set of experiments in which users of the same CPR decide on their individual resource harvesting rates to maximize their rewards, which in turn depends on the state of the resource that is evolving. We show that when rewards are given to players proportionally to the amount of resource they individually extract, self-interested behavior leads to overuse and depletion despite the greater long-term cumulated rewards of cooperative strategies. On the other hand cooperation may occur when individual decisions are informed by shared goals and changes in resource level. We finally then propose an analytical framework based on optimal control theory that is able to describe the collected data.
Samir Suweis is Assistant Professor (RTDB) at the University of Padova, Physics and Astronomy Department, member of the Padova Neuroscience Center (www.pnc.unipd.it) and the European Center of Living Systems (https://www.unive.it/pag/23664/). He co-lead the Laboratory of Interdisciplinary Physics (www.liphlab.com). His main research themes can be classified in three broad areas: 1) The formulation of simple principles to explain self-organization and emergent simplicity in nature; 2) Data analysis and complex network modeling and non-linear dynamics in socio-ecological systems; 3) Criticality in living systems, with a particular focus on brain criticality. In particular, his work focuses on the study of complex living systems under a theoretical framework provided by statistical mechanics. It addresses a wide range of related topics, including ecosystem organizations, ecological networks, stochastic modelling of ecosystems dynamics and eco-hydrological processes, sustainability and ecosystems services, brain networks and whole brain models. He served as member of the steering committee of the Complex System Society from 2018 to 2021 and he is vice-president of the Italian chapter of the Complex System Society.
Further info: https://suweis.github.io/ @SamirSuweis