Search

Resilience refers to the ability of systems to withstand, adapt to, and recover from disruptive events. While studies on resilience have attracted significant attention across various research domains, the precise definition of this concept within the field of cooperative artificial intelligence remains unclear. This paper addresses this gap by proposing a clear definition of `cooperative resilience' and outlining a methodology for its quantitative measurement. The methodology is validated in an environment with RL-based and LLM-augmented autonomous agents, subjected to environmental changes and the introduction of agents with unsustainable behaviors. These events are parameterized to create various scenarios for measuring cooperative resilience. The results highlight the crucial role of resilience metrics in analyzing how the collective system prepares for, resists, recovers from, sustains well-being, and transforms in the face of disruptions. These findings provide foundational insights into the definition, measurement, and preliminary analysis of cooperative resilience, offering significant implications for the broader field of AI. Moreover, the methodology and metrics developed here can be adapted to a wide range of AI applications, enhancing the reliability and effectiveness of AI in dynamic and unpredictable environments., Comment: Supplementary material in https://github.com/mavivi95/resilience/blob/main/Supplementary_File.pdf

As the field of AI continues to evolve, a significant dimension of this progression is the development of Large Language Models and their potential to enhance multi-agent artificial intelligence systems. This paper explores the cooperative capabilities of Large Language Model-augmented Autonomous Agents (LAAs) using the well-known Meltin Pot environments along with reference models such as GPT4 and GPT3.5. Preliminary results suggest that while these agents demonstrate a propensity for cooperation, they still struggle with effective collaboration in given environments, emphasizing the need for more robust architectures. The study's contributions include an abstraction layer to adapt Melting Pot game scenarios for LLMs, the implementation of a reusable architecture for LLM-mediated agent development - which includes short and long-term memories and different cognitive modules, and the evaluation of cooperation capabilities using a set of metrics tied to the Melting Pot's "Commons Harvest" game. The paper closes, by discussing the limitations of the current architectural framework and the potential of a new set of modules that fosters better cooperation among LAAs.

Social dilemmas are situations where groups of individuals can benefit from mutual cooperation but conflicting interests impede them from doing so. This type of situations resembles many of humanity's most critical challenges, and discovering mechanisms that facilitate the emergence of cooperative behaviors is still an open problem. In this paper, we study the behavior of self-interested rational agents that learn world models in a multi-agent reinforcement learning (RL) setting and that coexist in environments where social dilemmas can arise. Our simulation results show that groups of agents endowed with world models outperform all the other tested ones when dealing with scenarios where social dilemmas can arise. We exploit the world model architecture to qualitatively assess the learnt dynamics and confirm that each agent's world model is capable to encode information of the behavior of the changing environment and the other agent's actions. This is the first work that shows that world models facilitate the emergence of complex coordinated behaviors that enable interacting agents to ``understand'' both environmental and social dynamics., Comment: ICLR 2023 - AI4ABM workshop

Population games can be regarded as a tool to study the strategic interaction of a population of players. Although several attention has been given to such field, most of the available works have focused only on the unconstrained case. That is, the allowed equilibrium of the game is not constrained. To further extend the capabilities of population games, in this paper we propose a novel class of primal-dual evolutionary dynamics that allow the consideration of constraints that must be satisfied at the equilibrium of the game. Using duality theory and Lyapunov stability theory, we provide sufficient conditions to guarantee the asymptotic stability and feasibility of the equilibria set of the game under the considered constraints. Furthermore, we illustrate the application of the developed theory to some classical population games with the addition of constraints., Comment: 14 pages, 4 figures

High penetration of distributed generation will be characteristic to future distribution networks. The dynamic, intermittent, uncertain and deregulated nature of distributed generation raises the need for online, distributed economic dispatch techniques. In this paper, we demonstrate the application of such approaches using population dynamics. We propose a congestion management algorithm and demonstrate the notable properties and requirements of the proposed approach., Comment: Accepted for presentation at IEEE ISGT Europe 2020

While humans and animals learn incrementally during their lifetimes and exploit their experience to solve new tasks, standard deep reinforcement learning methods specialize to solve only one task at a time. As a result, the information they acquire is hardly reusable in new situations. Here, we introduce a new perspective on the problem of leveraging prior knowledge to solve future tasks. We show that learning discrete representations of sensory inputs can provide a high-level abstraction that is common across multiple tasks, thus facilitating the transference of information. In particular, we show that it is possible to learn such representations by self-supervision, following an information theoretic approach. Our method is able to learn concepts in locomotive and optimal control tasks that increase the sample efficiency in both known and unknown tasks, opening a new path to endow artificial agents with generalization abilities.

The modernization of the power system introduces technologies that may improve the system's efficiency by enhancing the capabilities of users. Despite their potential benefits, such technologies can have a negative impact. This subject has widely analyzed, mostly considering for-profit electric utilities. However, the literature has a gap regarding the impact of new technologies on non-profit utilities. In this work, we quantify the price of anarchy of non-profit utilities, that is, the cost caused by lack of coordination of users. We find that users, in the worst case, can consume up to twice the optimal demand, obtaining a small fraction of the optimal surplus. For this reason, we leverage the theory of mechanism design to design an incentive scheme that reduces the inefficiencies of the system, which preserves the privacy of users. We illustrate with simulations the efficiency loss of the system and show two instances of incentive mechanism that satisfy either budget balance and budget deficit., Comment: 17 pages, 3 figures, accepted in the International Journal of Electrical Power and Energy Systems

The vulnerability of false data injection attacks on real-time electricity pricing for the power grid market has been recently explored. Previous work has focused on the impact caused by attackers that compromise pricing signals and send false prices to a subset of consumers. In this paper we extend previous work by considering a more powerful and general adversary model, a new analysis method based on sensitivity functions, and by proposing several countermeasures that can mitigate the negative impact of these attacks. Countermeasures include adding a low-pass filter to the pricing signals, selecting the time interval between price updates, selecting parameters of the controller, designing robust control algorithms, and by detecting anomalies in the behavior of the system., Comment: 15 pages, 15 figures

In this work we investigate the inefficiency of the electricity system with strategic agents. Specifically, we prove that without a proper control the total demand of an inefficient system is at most twice the total demand of the optimal outcome. We propose an incentives scheme that promotes optimal outcomes in the inefficient electricity market. The economic incentives can be seen as an indirect revelation mechanism that allocates resources using a one-dimensional message space per resource to be allocated. The mechanism does not request private information from users and is valid for any concave customer's valuation function. We propose a distributed implementation of the mechanism using population games and evaluate the performance of four popular dynamics methods in terms of the cost to implement the mechanism. We find that the achievement of efficiency in strategic environments might be achieved at a cost, which is dependent on both the users' preferences and the dynamic evolution of the system. Some simulation results illustrate the ideas presented throughout the paper., Comment: 38 pages, 9 figures, submitted to journal

In Colombia, there is an increasing interest about improving public transportation. One of the proposed strategies in that way is the use battery electric vehicles (BEVs). One of the new challenges is the BEVs routing problem, which is subjected to the traditional issues of the routing problems, and must also consider the particularities of autonomy, charge and battery degradation of the BEVs. In this work, a scheme that coordinates the routing, scheduling of charge and operating costs of BEVs is proposed. The simplified operating costs have been modeled considering both charging fees and battery degradation. A case study is presented, in order to illustrate the proposed methodology. The given case considers an airport shuttle service scenario, in which energy consumption of the BEVs is estimated based on experimentally measured driving patterns., Comment: Highlights: We present a methodology for charging and route assignment for a BEVs fleet. We study the case of an airport shuttle service. We found the effects of scheduling of charge on the battery lifetime. Some conditions to perform a V2B operation are presented. V2B operation is only profitable under nonrealistic scenarios