CATALINA is an agent architecture conceived to extend a Practical Reasoner architecture inspired by Bratman's BDI paradigm with capacities that are central to agentive guidance: executive function, attention modulation, and the global availability of desire-relevant information [1, 2]. This paper discusses the CATALINA agent working cycle that exploits the desired architectural features and is an extension of the classical MAPE (Monitor-Analyse-Plan-Execute) loop. We also propose an implementation of our architecture and an experimental setup developed to test CATALINA's agent features in two different scenarios.
Multi-Agent Systems are characterized by multiple agents interacting to solve tasks that may be difficult, or even impossible, for a single agent. While discovering solutions to problems with a single objective might be relatively straightforward, the picture changes when coping with Multi-Objective Optimization (MOO), where problems require the simultaneous optimization of multiple objectives that potentially conflict with each other. This is particularly relevant in Multi-Agent Systems (MASs), since each agent's behavior affects the overall system performance. For example, the capability of a system, composed of many robots, to both locomote and aggregate simultaneously requires the definition of appropriate fitness measures and the usage of suitable algorithms. In this work, we investigate the conditions necessary to promote aggregation in a robotic MAS, with a particular focus on how conflicting objectives can hinder the learning of effective behaviors. Specifically, we designed a novel fitness function and tested it in a relatively simple aggregation scenario. Furthermore, we considered a recently introduced MOO problem, in which a MAS of five robots must develop the ability to aggregate while in motion. Our outcomes show that, despite the challenges in designing effective fitness functions, the proposed formulation successfully supports aggregation in the simpler scenario and enhances aggregation capabilities in the more complex one.
High availability (HA) systems are designed to guarantee availability of process and data for more then 99% of their operational time. HA are nowadays present in many contexts, both government and commercial services. In particular, the widespread of IoT smart devices has increased the need of HA systems providing broad network access. Many HA systems are built on top of the cloud, which is the standard de facto for small and medium companies to exploit reliable computational resources capable to scale up and down very quickly. Intelligent agents were developed more than 20 years ago, they have assumed a great importance in various context. Indeed, they are capable, in principle, to perceive the environment where they live and autonomously take actions to achieve goals in a wide range of contexts. Moreover, agents can improve their own performance by adopting machine learning techniques or by gaining knowledge about both environment and application domains. This position paper explores the use of intelligent agents to support core mechanisms-specifically monitoring, failure detection, and recovery-in HA systems. The discussion begins by reviewing key background concepts of HA architectures, followed by a structured characterization of their main components.
Human-machine interaction with large language models (LLMs) is built on an implicit trust in their ability to provide reliable, objective, and neutral information - an assumption that contrasts sharply with human-human interactions, where bias, conflict, and subjectivity naturally arise from embodied perspectives. Because LLMs are disembodied entities, they are often perceived as impartial and free from contradiction. This paper argues that such perceptions reflect a longstanding human aspiration: the utopian ideal of accessing "pure" knowledge-information unmediated by human subjectivity and as close to reality as possible. However, we challenge this assumption by demonstrating that bias and conflict remain structurally embedded within the data that LLMs process, reinterpret and generate. Rather than eliminating ambiguity, LLMs conceal it through a process of complexity reduction and an illusion of truth. Through a transdisciplinary analysis of LLM responses to culturally sensitive prompts, we reveal how ambiguity and conflict are systematically smoothed over in human-machine interactions. By examining empirical cases involving fine-tuning, dataset selection, and trigger-based interactions, we argue that LLMs are deliberately designed to produce responses that align with an idealized notion of 'universal humanity', a neutral, conflict-free, and harmonious representation of knowledge. This shaping of interactions reinforces a curated, utopian version of reality, influencing how users perceive and engage with AI-generated information.
