Autonomous Intelligent Agents Research Articles

Negotiation-based scheduling considering agent emotion

Scheduling in decentralized manufacturing systems is being expanded and transformed to a smart level where manufacturing tasks are executed autonomously. However, achieving self-organized manufacturing resources that can make collaborative decisions remains challenging. This paper presents a novel negotiation model for decentralized scheduling in an autonomous manufacturing system. A variety of emotions represent the willingness of autonomous intelligent agents (IAs) to negotiate for a consensus despite their diverse goals. The proposed communication protocol considers the emotional aspects of the autonomous machine agents in decentralized decision-making. The negotiation model focuses on analyzing the effects of two key emotions: selfishness and cooperation. This study constitutes the basis for experiments on multi-goal machine tasks and artificial emotion mechanism in a flow shop production system. The complexities of negotiation dynamics between IAs, where the interplay of selfish and cooperative strategies influences the outcomes and efficiency of reaching agreements, are explored. The experimental results show that selfish IAs eventually reach agreements through gradual compromise, but cooperative IAs, with their flexibility, can quickly achieve resolutions, especially when bargaining power is balanced. Aligning with the dual concern theory, this study reveals that while selfish IAs can dominate cooperative ones, the cooperative approach often leads to faster and more effective outcomes, regardless of the level of selfishness involved

Learning Fast and Slow: A Redux of Levels of Learning in General Autonomous Intelligent Agents

Autonomous intelligent agents, including humans, operate in a complex, dynamic environment that necessitates continuous learning. We revisit our thesis that proposes that learning in human-like agents can be categorized into two levels: Level 1 (L1) involving innate and automatic learning mechanisms, while Level 2 (L2) comprises deliberate strategies controlled by the agent. Our thesis draws from our experiences in building artificial agents with complex learning behaviors, such as interactive task learning and open-world learning.

Turing-like Experiment in a Cyber Defense Game

During the past decade, researchers of behavioral cyber security have created cognitive agents that are able to learn and make decisions in dynamic environments in ways that assimilate human decision processes. However, many of these efforts have been limited to simple detection tasks and represent basic cognitive functions rather than a whole set of cognitive capabilities required in dynamic cyber defense scenarios. Our current work aims at advancing the development of cognitive agents that learn and make defense-dynamic decisions during cyber attacks by intelligent attack agents. We also aim to evaluate the capability of these cognitive models in ``Turing-like'' experiments, comparing the decisions and performance of these agents against human cyber defenders. In this paper, we present an initial demonstration of a cognitive model of the defender that relies on a cognitive theory of dynamic decision-making, Instance-Based Learning Theory (IBLT); we also demonstrate the execution of the same defense task by human defenders. We rely on OpenAI Gym and CybORG and adapt an existing CAGE scenario to generate a simulation experiment using an IBL defender. We also offer a new Interactive Defense Game (IDG), where \textit{human} defenders can perform the same CAGE scenario simulated with the IBL model. Our results suggest that the IBL model makes decisions against two intelligent attack agents that are similar to those observed in a subsequent human experiment. We conclude with a description of the cognitive foundations required to build autonomous intelligent cyber defense agents that can collaborate with humans in autonomous cyber defense teams.

Technologies of Bioimpedance Spectroscopy in Decision Support Systems for the Diagnosis of Socially Significant Diseases

The purpose of the research is to develop methods for the synthesis of hybrid classifiers to assess the risk of socially significant diseases using bioimpedance analysis.Methods. We developed a descriptor approach using impedance spectroscopy results, generating four amplitudephase-frequency responses from four quasi-orthogonal leads. They create the feature spaces necessary for our hybrid classifier in the diagnosis of pancreatic diseases, the autonomous intelligent agents of which are built on various paradigms: probabilistic neural networks, fuzzy logical inference, fully connected feedforward neural networks. We also presented a device structure for creating an informative feature space.Results. Experimental studies of the proposed methods and means of classifying medical rice were carried out on diagnostic tasks according to the classes "acute destructive pancreatitis" – "no acute destructive pancreatitis" and differential diagnosis tasks according to the classes "prostate cancer" ‒ "chronic pancreatitis". They showed that incorporating multi-frequency sensing into neural network-based classifiers allows the development of clinical decision support systems for disease diagnosis that are comparable in performance to existing clinical diagnostic methods. The results were confirmed in groups of male and female patients at different stages of cancer aged 25 to 80 years using a variety of diagnostic methods, including history, physical examination, assessment of comorbidities, laboratory tests, ultrasound, laparoscopy, intraoperative exploration and computed tomography.Conclusion. The use of bioimpedance spectroscopy and hybrid classifier models opens up new opportunities for accessible and objective diagnosis of pancreatic diseases, expanding the capabilities of intelligent medical decision support systems.

The Observable Mind: Enabling an Autonomous Agent Sharing Its Conscious Contents Using a Cognitive Architecture

We enable an autonomous agent sharing its artificial mind to its audiences like humans. This supports the autonomous human robot interactions relying on a cognitive architecture, LIDA, which explains and predicts how minds work and is used as the controllers of intelligent autonomous agents. We argue that LIDA’s cognitive representations and processes may serve as the source of the mind content its agent shares out, autonomously. We proposed a new description (sub) model into LIDA, letting its agent describing its conscious contents. Through this description, the agent’s mind is more observable so we can understand the agent’s entity and intelligence more directly. Also, this helps the agent explains its behaviors to its audiences so engage into its living society better. We built an initial LIDA agent embedding with this description model. The agent shares its conscious content autonomously, reasonably explaining its behaviors.

Reinforcement Learning as a Path to Autonomous Intelligent Cyber-Defense Agents in Vehicle Platforms

Technological advancement of vehicle platforms exposes opportunities for new attack paths and vulnerabilities. Static cyber defenses can help mitigate certain attacks, but those attacks must generally be known ahead of time, and the cyber defenses must be hand-crafted by experts. This research explores reinforcement learning (RL) as a path to achieve autonomous, intelligent cyber defense of vehicle control networks—namely, the controller area network (CAN) bus. We train an RL agent for the CAN bus using Toyota’s Portable Automotive Security Testbed with Adaptability (PASTA). We then apply the U.S. Army Combat Capabilities Development Command (DEVCOM) Army Research Laboratory’s methodology for quantitative measurement of cyber resilience to assess the agent’s effect on the vehicle testbed in a contested cyberspace environment. Despite all defenses having similar traditional performance measures, our RL agent averaged a 90% cyber resilience measurement during drive cycles executed on hardware versus 41% for a naïve static timing defense and 98% for the bespoke timing-based defense. Our results also show that an RL-based agent can detect and block injection attacks on a vehicle CAN bus in a laboratory environment with greater cyber resilience than prior learning approaches (1% for convolutional networks and 0% for recurrent networks). With further research, we believe there is potential for using RL in the autonomous intelligent cyber defense agent concept.

Modeling Theory of Mind in Dyadic Games Using Adaptive Feedback Control

A major challenge in cognitive science and AI has been to understand how intelligent autonomous agents might acquire and predict the behavioral and mental states of other agents in the course of complex social interactions. How does such an agent model the goals, beliefs, and actions of other agents it interacts with? What are the computational principles to model a Theory of Mind (ToM)? Deep learning approaches to address these questions fall short of a better understanding of the problem. In part, this is due to the black-box nature of deep networks, wherein computational mechanisms of ToM are not readily revealed. Here, we consider alternative hypotheses seeking to model how the brain might realize a ToM. In particular, we propose embodied and situated agent models based on distributed adaptive control theory to predict the actions of other agents in five different game-theoretic tasks (Harmony Game, Hawk-Dove, Stag Hunt, Prisoner’s Dilemma, and Battle of the Exes). Our multi-layer control models implement top-down predictions from adaptive to reactive layers of control and bottom-up error feedback from reactive to adaptive layers. We test cooperative and competitive strategies among seven different agent models (cooperative, greedy, tit-for-tat, reinforcement-based, rational, predictive, and internal agents). We show that, compared to pure reinforcement-based strategies, probabilistic learning agents modeled on rational, predictive, and internal phenotypes perform better in game-theoretic metrics across tasks. The outlined autonomous multi-agent models might capture systems-level processes underlying a ToM and suggest architectural principles of ToM from a control-theoretic perspective.

Choice of conceptualization of a problem situation by an intelligent agent in decision-making tasks

The research in the domain of autonomous intelligent agent is the foreground of the introduction of artificial intelligence solution in all areas of economy. The intelligent autonomous systems combine the usage of pattern recognition, reasoning, decision making, conceptual modeling techniques and methods. The important part of intelligent agent implementation is to find the conceptualization which is suitable to the current problematic situation. Despite all progress around autonomous intelligent agents, humans are much more flexible and creative in making the right conceptualizations. They seamlessly adapt to the situation at hand and filter out all irrelevant details, using multiple perspectives and representations for the same objects. This research makes assumptions that every intelligent agent dynamically creates its own ontology used to interpret local knowledge. The mappings are established with this local ontology and the ontologies of other agents when needed, in order to share and reuse knowledge. In the article a formal model of problematic situation in the context of decision-making operation is presented. Models used in decision making and their relationships are described. In the second part of the article we analyze the process of conceptualization selection and arrive to the conclusion that this selection is done on multiple levels, starting from selecting the communicating agent with relevant domain of expertise, selecting and aligning ontologies of agents, selecting patterns and patterns languages which better correspond to the situation and lastly, selecting the relevant interpretations of concepts and relationships. In the last part of article, the problem of the selection of relevant knowledge provider is solved, using modified TOPSIS method. The proposed approach and directions of research will help to add flexibility to conceptual modeling of problematic situations by intelligent agents.

Enabling an autonomous agent sharing its minds, describing its conscious contents

We enable an autonomous agent (robot) to share its artificial mind, particularly its conscious content, with its audiences like humans or other agents. This effort is relying on the studies from cognitive modeling and supports the autonomous human robot interaction. LIDA is a biologically inspired systems-level cognitive model that explains and predicts how minds work. This model has been used in designing the controllers of intelligent autonomous agents. Here we propose to add a new description (sub) model into LIDA, allowing its agents to share their control structures, the artificial minds. We argue that the cognitive representations and processes introduced in LIDA may naturally serve as the source of the mind contents an agent can express out. We also proposed a type of description behavior based on LIDA’s action modules, which support the interactions between the agent and its real-world environments including human audiences.Through this description model, an agent’s mind will be more visible and accessible, so the agent’s intelligence can be assessed from not only its outside appearance and behaviors, but also its mind. In addition, the agent’s mind activities may associate with, so help explain certain behaviors it acted/acts. This helps the agent earn more understanding from its audiences so to being accepted by and engage to its living society better. Finally, as an agent’s controller, LIDA is designed heavily relying on the hypotheses of human minds, so its agent’s artificial mind contents are conceptually easy to understand to its human audiences.

Development in intelligent autonomous agents for the high-level cognitive functions like reasoning, planning, learning and abstraction

Intelligent Autonomous Agents (IAAs) is basic to the endurance and improvement of the firm, making new items as a wellspring of supportable upper hand. Based on Information-based hypothesis, this paper centers around the four measurements that are obtaining, processing, change double-dealing and investigated the connection between absorptive limit and item advancement execution from the possible absorptive limit and acknowledged absorptive limit. This paper is useful to clarify the irregularity between absorptive limit and item advancement execution in surviving exploration, and further uncover the system between absorptive limit and new item development execution, enhance, extend the absorptive limit hypothesis and item advancement hypothesis by joining the two speculations. This paper can assist firms with explaining the connection between potential absorptive limit acknowledged absorptive limit and development execution, and give reference to firm dynamic, and can likewise give direction viably to firms to dispense assets.

Reformulation techniques for automated planning: a systematic review

Abstract Automated planning is a prominent area of Artificial Intelligence and an important component for intelligent autonomous agents. A cornerstone of domain-independent planning is the separation between planning logic, that is the automated reasoning side, and the knowledge model, that encodes a formal representation of domain knowledge needed to reason upon a given problem to synthesize a solution plan. Such a separation enables the use of reformulation techniques, which transform how a model is represented in order to improve the efficiency of plan generation. Over the past decades, significant research effort has been devoted to the design of reformulation techniques. In this paper, we present a systematic review of the large body of work on reformulation techniques for classical planning, aiming to provide a holistic view of the field and to foster future research in the area. As a tangible outcome, we provide a qualitative comparison of the existing classes of techniques, that can help researchers gain an overview of their strengths and weaknesses.

Semantic 3D city interfaces—Intelligent interactions on dynamic geospatial knowledge graphs

Abstract This article presents a system architecture and a set of interfaces that can build scalable information systems capable of large city modeling based on dynamic geospatial knowledge graphs to avoid pitfalls of Web 2.0 applications while blending artificial and human intelligence during the knowledge enhancement processes. We designed and developed a GeoSpatial Processor, an SQL2SPARQL Transformer, and a geospatial tiles ordering tasks and integrated them into a City Export Agent to visualize and interact with city models on an augmented 3D web client. We designed a Thematic Surface Discovery Agent to automatically upgrade the model’s level of detail to interact with thematic parts of city objects by other agents. We developed a City Information Agent to help retrieve contextual information, provide data concerning city regulations, and work with a City Energy Analyst Agent that automatically estimates the energy demands for city model members. We designed a Distance Agent to track the interactions with the model members on the web, calculate distances between objects of interest, and add new knowledge to the Cities Knowledge Graph. The logical foundations and CityGML-based conceptual schema used to describe cities in terms of the OntoCityGML ontology, together with the system of intelligent autonomous agents based on the J-Park Simulator Agent Framework, make such systems capable of assessing and maintaining ground truths with certainty. This new era of GeoWeb 2.5 systems lowers the risk of deliberate misinformation within geography web systems used for modeling critical infrastructures.

Research on decentralized autonomous fault processing of distribution network based on data twin Technology

With the development of new power system construction, distribution network as an important part, its safe and stable operation is more important. In order to make the distribution network have self-identification, self-positioning and self-healing functions, it is necessary to study the decentralized autonomous technology of the distribution network, including the state assessment and fault location technology of distributed power supply. In this paper, the autonomous intelligent agent unit (IA) is designed based on digital twin technology, which has the functions of state assessment, fault location and isolation. Through IEEE33-node with distributed power supply and MA-PSO algorithm, the internal autonomy of IA is simulated and analyzed. The results show that the proposed scheme and algorithm are fast, accurate and referable.

Multi-Agent Interaction to Assist Visually-Impaired and Elderly People.

A voice-controlled smart home system based on conversational agents can address the specific needs of older people, proactively providing support, compensating for cognitive decline, and coping with solitude, among other features. In particular, Multi-Agent Systems (MAS) platforms provide considerable support for complex adaptive systems that are naturally distributed and situated in dynamic environments, such as Ambient intelligence (AmI) applications. Such autonomous intelligent agents are capable of independent reasoning and joint analysis of complex situations to support high-level interaction with humans, besides providing typical characteristics of MAS, such as cooperation and coordinated action. In this context, we developed an approach using a MAS previously evaluated for visually impaired users, where most of the system’s functionalities are also helpful for the elderly. Our methodology is based on the four steps of the interactive design process. As a result, we determined that our approach has elements that allow for natural interaction with users, and we identified and discussed improvements and new features for future work. We believe that our findings can point to directions for building AmI systems that are capable of more natural interaction with users.

Characterisation of Intelligent Autonomous Agents Inspired by Biological Theory in Cognitive Environment

The paper investigates a naturally motivated meaning of wise shrewd self-ruling specialists of humans. Knowledge is identified with whether the conduct of a framework adds to its self-upkeep. Conduct turns out to be clearer (or adapts to more biological issue factors) when it is able to make and utilize portrayals. The thought of portrayal ought not to be confined to formal articulations with neuro hypothetical semantics. The element at different degrees of canny frameworks assumes a fundamental part in shaping portrayals. The paper investigates an organically roused meaning of shrewd self-ruling specialists of humans. Insight is identified with whether the conduct of a framework adds to its self-upkeep. Conduct turns out to be keener when it is proficient to make and utilize portrayals. The idea of portrayal also focused on formal articulations with sentimental hypothetical semantics.

Hierarchical intrinsically motivated agent planning behavior with dreaming in grid environments

Biologically plausible models of learning may provide a crucial insight for building autonomous intelligent agents capable of performing a wide range of tasks. In this work, we propose a hierarchical model of an agent operating in an unfamiliar environment driven by a reinforcement signal. We use temporal memory to learn sparse distributed representation of state–actions and the basal ganglia model to learn effective action policy on different levels of abstraction. The learned model of the environment is utilized to generate an intrinsic motivation signal, which drives the agent in the absence of the extrinsic signal, and through acting in imagination, which we call dreaming. We demonstrate that the proposed architecture enables an agent to effectively reach goals in grid environments.

Semantic 3D City Agents—An intelligent automation for dynamic geospatial knowledge graphs

This paper presents a system of autonomous intelligent software agents, based on a cognitive architecture, capable of automated instantiation, visualisation and analysis of multifaceted City Information Models in dynamic geospatial knowledge graphs. Design of JPS Agent Framework and Routed Knowledge Graph Access components was required in order to provide backbone infrastructure for an intelligent agent system as well as technology agnostic knowledge graph access enabling automation of multi-domain data interoperability. Development of CityImportAgent, CityExportAgent and DistanceAgent showcased intelligent automation capabilities of the Cities Knowledge Graph. The agents successfully created a semantic model of Berlin in LOD 2, compliant with CityGML 2.0 standard and consisting of 419 909 661 triples described using OntoCityGML. The system of agents also visualised and analysed the model by autonomously tracking interactions with a web interface as well as enriched the model by adding new information to the knowledge graph. This way it was possible to design a geospatial information system able to meet demands imposed by the Industry 4.0 and link it with the other multi-domain knowledge representations of The World Avatar.

Social identity theory applied to the game of self-regulation of social exchanges based on multiagent systems

ABSTRACT Autonomous intelligent agents participate in complex social situations. It is necessary that agents are aware of their society and still understand their context of action and interaction. Such social situation with other agents is influenced by society’s characteristics, including their norms, values, common interests and objectives, members, and social structures. Each complex society is broken down into social groups, each one with an identity. Social identity theory is seen as an analysis of intergroup relations in social categories based on the cognitive definition and the autoconcept of a social group and their belonging. Researchers have been studying this theory in computing, more specifically in multiagent systems. Social exchanges are the object of study in diverse contexts in which social relations are understood as social exchanges. One of the investigated themes is the self-regulation of social exchanges. The game of self-regulation of social exchanges explores its self-regulation but it does not analyse the agents’ identity. Therefore, this research investigates the integration of psychosocial analysis of social identity theory into the social adaptive complex society of the game of self-regulation of social exchanges. The principal result of this work is the GSRSEPSIT, a social simulation model.

Elements of Conceptual Thinking in Planning the Behavior of Autonomous Intelligent Agents

The expediency of using the tools of visual-effective, visual-figurative and conceptual thinking for planning the purposeful activity of autonomous intelligent agents in problem environments of various degrees of a priori uncertainty has been substantiated. The content is revealed and the role of each form of thinking is shown in the process of automatic planning of the purposeful behavior of autonomous intelligent agents in the changing conditions of functioning. The special role of conceptual thinking in the performance of complex tasks by autonomous agents and the planning of polyphasic behavior associated with it is indicated. Taking into account the complexity of the problems associated with the formalization of mental acts of conceptual thinking, possible ways of its gradual development from the initial level to the transition to higher levels of development are shown, expanding on this basis the class of tasks solved by autonomous intelligent agents. A model of knowledge representation and tools for deriving solutions of the initial level of conceptual thinking have been developed, which allow intelligent agents to break down the tasks they receive into sub-goals of behavior. Then, on this basis, plan polyphase activity by searching for solutions to the associated subtasks, which ensure the determination of the minimum length routes of movement in a prob lematic environment with obstacles and the purposeful manipulation of objects in it. The tools are synthesized allowing to establish the order of elaboration of complex actions included in the structure of the task formulated by autonomous intelligent agents. It is shown that the further development of the proposed methodological foundations for constructing intelligent problem sol vers is associated with the formalization of a higher level of mental acts of conceptual thinking, which make it possible to solve practical problems of different complexity, formulated both in procedural and declarative form of presentation in the form of various target situations of the problem environment, having a large dimension.

Intelligent autonomous agents and trust in virtual reality

Intelligent autonomous agents (IAA) are proliferating and rapidly evolving due to the exponential growth in computational power and recent advances, for instance, in artificial intelligence research. Ranging from chatbots, over personal virtual assistants and medical decision-aiding systems, to self-driving or self-piloting systems, whether unbeknownst to the users or not, IAA are increasingly integrated into many aspects of daily life. Despite this technological development, many people remain skeptical of such agents. Conversely, others might have excessive confidence in them. Therefore, establishing an appropriate level of trust is crucial to the successful deployment of IAA in everyday contexts. Virtual Reality (VR) is another domain where IAA play a significant role, yet its experiential and immersive character particularly allows for new ways of interaction and tackling trust-related issues. In this article, we provide an overview of the numerous factors involved in establishing trust between users and IAA, spanning scientific disciplines as diverse as psychology, philosophy, sociology, computer science, and economics. Focusing on VR, we discuss the different types and definitions of trust and identify foundational factors classified into three interrelated dimensions: Human-Technology, Human-System, and Interpersonal. Based on this taxonomy, we identify open issues and a research agenda towards facilitating the study of trustful interaction and collaboration between users and IAA in VR settings.