Search

Integrating large language model (LLM) agents within game theory demonstrates their ability to replicate human-like behaviors through strategic decision making. In this paper, we introduce an augmented LLM agent, called the private agent, which engages in private deliberation and employs deception in repeated games. Utilizing the partially observable stochastic game (POSG) framework and incorporating in-context learning (ICL) and chain-of-thought (CoT) prompting, we investigated the private agent's proficiency in both competitive and cooperative scenarios. Our empirical analysis demonstrated that the private agent consistently achieved higher long-term payoffs than its baseline counterpart and performed similarly or better in various game settings. However, we also found inherent deficiencies of LLMs in certain algorithmic capabilities crucial for high-quality decision making in games. These findings highlight the potential for enhancing LLM agents' performance in multi-player games using information-theoretic approaches of deception and communication with complex environments. [ABSTRACT FROM AUTHOR]

Testing of a software product is the key activity before deploying it in the real-time environment. Therefore testing should be introduced into the product development process as early as possible in order to decrease the costs of repairing the damage in later phases. In this paper we use Coloured Petri Nets (CPN) for the system specification and also as a system to be tested instead of its implementation. This specification also serves as the basis for the test suite generation. Test cases are provided in the language of Testing and Test Control Notation version 3 (TTCN-3) due to its general application area and platform independence. Stop-and-wait communication protocol has been used as a tutorial example of our approach. [ABSTRACT FROM AUTHOR]

Confirmed cases of the disease COVID-19 have spread to more than 200 countries and regions of the world within a few months. Although the authorities report the number of new cases on daily basis, there remains a gap between the number of reported cases and actual number of cases in a population. One way to bridge this gap is to gain more in-depth understanding of the disease. In this paper, we have used the recent findings about the clinical courses of inpatients with COVID-19 to reset the initial conditions of the epidemic process in order to estimate more realistic number of cases in the population. By translating the reported cases certain number of days earlier with regard to an average clinical course of the disease, we have obtained much higher number of cases, which suggests that the actual number of infected cases and death rate might have been higher than reported. Based on the outbreak of COVID-19 in Italy, this paper shows an estimate of the number of infected cases based on infection and removal rates from data during the pandemic., (This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.)