Learning algorithm for an intelligent decision making system based on multi-agent neurocognitive architectures

The paper presents the formalism of an intelligent decision-making system based on multi-agent neurocognitive architectures, which has an architectural similarity to the human brain. An invariant of the organizational and functional structure of the intellectual decision-making process based on the multi-agent neurocognitive architecture is developed. An algorithm for teaching intelligent decision-making systems based on the self-organization of the invariant of multi-agent neurocognitive architectures is presented. Using this algorithm, an intelligent agent was trained and the architecture of the learning process was built on the basis of an invariant of neurocognitive architecture. Further research is related to training an intelligent agent in more complex behavior and expanding the capabilities of an intelligent decision-making system based on multi-agent neurocognitive architectures.     

Learning data-driven decision-making policies in multi-agent environments for autonomous systems

Autonomous systems such as Connected Autonomous Vehicles (CAVs), assistive robots are set improve the way we live. Autonomous systems need to be equipped with capabilities to Reinforcement Learning (RL) is a type of machine learning where an agent learns by interacting with its environment through trial and error, which has gained significant interest from research community for its promise to efficiently learn decision making through abstraction of experiences. However, most of the control algorithms used today in current autonomous systems such as driverless vehicle prototypes or mobile robots are controlled through supervised learning methods or manually designed rule-based policies. Additionally, many emerging autonomous systems such as driverless cars, are set in a multi-agent environment, often with partial observability. Learning decision making policies in multi-agent environments is a challenging problem, because the environment is not stationary from the perspective of a learning agent, and hence the Markov properties assumed in single agent RL does not hold. This paper focuses on learning decision-making policies in multi-agent environments, both in cooperative settings with full observability and dynamic environments with partial observability. We present experiments in simple, yet effective, new multi-agent environments to simulate policy learning in scenarios that could be encountered by an autonomous navigating agent such as a CAV. The results illustrate how agents learn to cooperate in order to achieve their objectives successfully. Also, it was shown that in a partially observable setting, an agent was capable of learning to roam around its environment without colliding in the presence of obstacles and other moving agents. Finally, the paper discusses how data-driven multi-agent policy learning can be extended to real-world environments by augmenting the intelligence of autonomous vehicles.     

A Sat-Based Approach to Unbounded Model Checking for Alternating-Time Temporal Epistemic Logic

This paper deals with the problem of verification of game-like structures by means of symbolic model checking. Alternating-time Temporal Epistemic Logic (ATEL) is used for expressing properties of multi-agent systems represented by alternating epistemic temporal systems as well as concurrent epistemic game structures. Unbounded model checking (a SAT based technique) is applied for the first time to verification of ATEL. An example is given to show an application of the technique.     

Proof-Theoretic Analysis of the Logics of Agency: The Deliberative STIT

Studia Logica - A sequent calculus methodology for systems of agency based on branching-time frames with agents and choices is proposed, starting with a complete and cut-free system for multi-agent...     

社会群体中的偏好逻辑

在许多多主体偏好逻辑系统中,主体之间是没有联系的,因而无法描述主体间的偏好互动。借鉴＂The Logic in the Community＂一文中称为＂群体压力＂的例子对偏好的影响,本文在＂The Logic in the Community＂所提出的系统构架上,加入偏好算子和两个动态算子,通过构造归约公理说明这些算子可以被无偿添加,并借助这些规约公理证明了系统的完全性。     

Multi-Agent neurocognitive models of semantics of spatial localization of events

The purpose of the study is to develop a learning system for internal representation of the events localization space to realize orientation and navigation of autonomous mobile systems. The task of the research is the development of simulation models of the semantics of the event localization space based on multi-agent neurocognitive architectures. The paper proves that the multi-agent neurocognitive architecture is an effective formalism for describing the semantics of the spatial localization of events. Main theoretical foundations have been developed for the simulation of spatial relations using the so-called multi-agent facts, consisting of software agents-concepts, reflecting semantic categories corresponding to parts of speech. It is shown that locative software agents that describe the spatial location of objects and events, forming homogeneous connections, compose the so-called field locations. The latter describes a holistic view of the intellectual agent about the environment. The paper defines conceptual foundations of multi-agent modeling of the semantics of subjective reflexive mapping of the interaction between real objects, space and time.     

Automatic verification of multi-agent systems by model checking via ordered binary decision diagrams

We present a methodology for the verification of multi-agent systems, whose properties are specified by means of a modal logic that includes a temporal, an epistemic, and a modal operator to reason about correct behaviour of agents. The verification technique relies on model checking via ordered binary decision diagrams. We present an implementation and report on experimental results for two scenarios: the bit transmission problem with faults and the protocol of the dining cryptographers.     

Learning HMM-based cognitive load models for supporting human-agent teamwork

Cognitive studies indicate that members of a high performing team often develop shared mental models to predict others’ needs and coordinate their behaviors. The concept of shared mental models is especially useful in the study of human-centered collaborative systems that require humans to team with autonomous agents in complex activities. We take the position that in a mixed human-agent team, agents empowered with cognitive load models of human team members can help humans develop better shared mental models to enhance team performance. Inspired by human information processing system, we here propose a HMM-based load model for members of human-agent teams, and investigate the development of realistic cognitive load models. A cognitive experiment was conducted in team contexts to collect data about the observable secondary task performance of human participants. The data were used to train hidden Markov models (HMM) with varied numbers of hypothetical hidden states. The result indicates that the model spaces have a three-layer structure. Statistical analysis also reveals some characteristics of the models at the top-layer. This study can be used in guiding the selection of HMM-based cognitive load models for agents in human-centered multi-agent systems.     

Performance structures: A psycholinguistic and linguistic appraisal

Two lines of research—one in psycholinguistics and one in linguistics—are combined to deal with a long-standing problem in both fields: why the “performance structures” of sentences (structures based on experimental data, such as pausing and parsing values) are not fully accountable for by linguistic theories of phrase structure. Two psycholinguistic algorithms that have been used to predict these structures are described and their limitations are examined. A third algorithm, based on the prosodic structures of sentences is then proposed and shown to be a far better predictor of performance structures. It is argued that the experimental data reflect aspects of the linguistic cognitive capacity, and that, in turn, linguistic theory can offer an illuminating account of the data. The prosodic model is shown to have a wider domain of application than temporal organization per se, accounting for parsing judgments as well as pausing performance, and reflecting aspects of syntactic and semantic structure as well as purely prosodic structure. Finally, the algorithm is discussed in light of language processing.     

Agent communication pragmatics: the cognitive coherence approach

Different approaches have investigated the syntax and semantics of agent communication languages. However, these approaches have not indicated how agents should dynamically use communications. Instead of filling this pragmatics gap, most approaches have mainly focused on the ‘structure’ of dialogues even though developers are more interested in agents’ capabilities of having ‘useful’ automated conversations with respect to their goals rather than in their abilities to structure dialogues. This led us to work on a theory of the use of conversations between agents. In this paper, we propose a pragmatics theory which extends and adapts the cognitive dissonance theory (a major theory of social psychology) to multi-agent systems by unifying it with the theory of coherence in thought and action that issues from computational philosophy of mind. Precisely, we show how this theory allows us to provide generic conceptual tools for the automation of both agent communicational behavior and attitude change processes. This new motivational model is formulated in terms of constraints and elements of cognition and allows us to define cognitive incoherences and dialogue utility measures. We show how these measures could be used to solve common problems and answer some critical questions concerning agent communication frameworks use. Finally, our exploration in applying the cognitive coherence pragmatics theory as a new communication layer over classical BDI agents is presented. It relies on our dialogue games based agent communication language (DIAGAL) and our dialogue games simulator toolbox (DGS). The resulting framework provides the necessary theoretical and practical elements for implementing our theory. In doing so, it brings in a general scheme for automatizing agents’ communicational behavior as it is exemplified in this article.     

Subjective Situations and Logical Omniscience

The beliefs of the agents in a multi-agent system have been formally modelled in the last decades using doxastic logics. The possible worlds model and its associated Kripke semantics provide an intuitive semantics for these logics, but they commit us to model agents that are logically omniscient. We propose a way of avoiding this problem, using a new kind of entities called subjective situations. We define a new doxastic logic based on these entities and we show how the belief operators have some desirable properties, while avoiding logical omniscience. A comparison with two well-known proposals (Levesque's logic of explicit and implicit beliefs and Thijsse's hybrid sieve systems) is also provided.     

Security of multi-agent systems: A case study on comparison shopping

The multi-agent-systems paradigm is becoming more and more popular as a basis for realizing net-based solutions. This development is accompanied by an increasing relevance of security issues. For instance, the potential loss of privacy and other assets is a major concern for, both merchants and customers, in Internet-based commerce and, without being properly addressed, such very legitimate concerns hamper the growth of e-commerce.This article uses a comparison-shopping scenario to introduce a general methodology for formally verifying the security of multi-agent systems. Following the approach of possibilistic information flow security, the flow of information between and within agents is restricted in order to ensure that secrets will not be disclosed to unauthorized meddlers. The security requirements for the overall system are then decomposed into requirements for the individual agents that can be verified independently from each other. Exploiting the modular structure of a multi-agent system considerably reduces the complexity of the overall security analysis. The techniques for decomposing security requirements, for verifying individual agents, and for deriving global security guarantees for the entire system from locally verified properties are all generic in the sense that they apply also to many other systems and security requirements than the ones that appear in the example scenario.     

A formalisation of violation, error recovery, and enforcement in the bit transmission problem

The design of complex multi-agent systems is increasingly having to confront the possibility that agents may not behave as they are supposed to. In addition to analysing the properties that hold if protocols are followed correctly, it is also necessary to predict, test, and verify the properties that would hold if these protocols were to be violated. We illustrate how the formal machinery of deontic interpreted systems can be applied to the analysis of such problems by considering three variations of the bit transmission problem. The first, an example in which an agent may fail to do something it is supposed to do, shows how we deal with violations of protocols and specifications generally. The second, an example in which an agent may do something it is not supposed to do, shows how it is possible to specify and analyse remedial or error-recovery procedures. The third combines both kinds of faults and introduces a new component into the system, a controller, whose role is to enforce compliance with the protocol. In each case the formal analysis is used to test whether critical properties of the system are compromised, in this example, the reliable communication of information from one agent to the other.     

Integrating a cognitive assistant within a critique-based recommender system

Recommender systems are cognitive computing systems designed to support humans in their decision-making processes through convincing, timely product suggestions. In the field of recommender systems, critique-based recommenders have been widely applied as an effective approach for guiding users through a product space in pursuit of suitable products. To date, no critique-based approach has included an assistant that support users in their search in a pleasant way. In this paper, we describe how we integrate an assistant within a critique-based recommender. We consider the proposed assistant to be cognitive because its reasoning process when recommending products is based on a cognitively-inspired clustering algorithm. The proposal is evaluated by users and compared with a non-assistant approach. The results of this research demonstrate that the integration of a cognitive assistant within the recommender improves the user experience and increases the performance of the recommendation process, i.e., users need fewer cycles to achieve the desired product or service.     

An analytic tableau calculus for a temporalised belief logic

A tableau is a refutation-based decision procedure for a related logic, and is among the most popular proof procedures for modal logics. In this paper, we present a labelled tableau calculus for a temporalised belief logic called TML⁺, which is obtained by adding a linear-time temporal logic onto a belief logic by the temporalisation method of Finger and Gabbay. We first establish the soundness and the completeness of the labelled tableau calculus based on the soundness and completeness results of its constituent logics. We then sketch a resolution-type proof procedure that complements the tableau calculus and also propose a model checking algorithm for TML⁺ based on the recent results for model checking procedures for temporalised logics. TML⁺ is suitable for formalising trust and agent beliefs and reasoning about their evolution for agent-based systems. Based on the logic TML⁺, the proposed labelled tableau calculus could be used for analysis, design and verification of agent-based systems operating in dynamic environments.     

Substantive and procedural norms in normative multiagent systems

Procedural norms are instrumental norms addressed to agents playing a role in the normative system, for example to motivate these role playing agents to recognize violations or to apply sanctions. Procedural norms have first been discussed in law, where they address legal practitioners such as legislators, lawyers and policemen, but they are discussed now too in normative multiagent systems to motivate software agents. Procedural norms aim to achieve the social order specified using regulative norms like obligations and permissions, and constitutive norms like counts-as obligations. In this paper we formalize procedural, regulative and constitutive norms using input/output logic enriched with an agent ontology and an abstraction hierarchy. We show how our formalization explains Castelfranchi's notion of mutual empowerment, stating that not only the agents playing a role in a normative system are empowered by the normative system, but the normative system itself is also empowered by the agents playing a role in it. In our terminology, the agents are not only institutionally empowered, but they are also delegated normative goals from the system. Together, institutional empowerment and normative goal delegation constitute a mechanism which we call delegation of power, where agents acting on behalf of the normative system become in charge of recognizing which institutional facts follow from brute facts.     

Term-Modal Logics

Many powerful logics exist today for reasoning about multi-agent systems, but in most of these it is hard to reason about an infinite or indeterminate number of agents. Also the naming schemes used in the logics often lack expressiveness to name agents in an intuitive way.To obtain a more expressive language for multi-agent reasoning and a better naming scheme for agents, we introduce a family of logics called term-modal logics. A main feature of our logics is the use of modal operators indexed by the terms of the logics. Thus, one can quantify over variables occurring in modal operators. In term-modal logics agents can be represented by terms, and knowledge of agents is expressed with formulas within the scope of modal operators.This gives us a flexible and uniform language for reasoning about the agents themselves and their knowledge. This article gives examples of the expressiveness of the languages and provides sequent-style and tableau-based proof systems for the logics. Furthermore we give proofs of soundness and completeness with respect to the possible world semantics.     

The performance space measurement of regional innovation system based on neuropsychology

In this paper, the performance space measurement of regional innovation system was studied based on neuropsychology. Firstly, the neuropsychology and neural evolution theory were elaborated. Secondly, the genetic algorithm was used to design a regional enterprise performance space measurement model, and this was obtained by connecting the ERP production module and RBF neural network order forecast module. Finally, the algorithm and model constructed in this paper were used to predict the performance of regional foreign trade innovation system. Then, it is concluded that the model constructed in this paper includes the network with the lowest network structure complexity, the smallest training error and the least test error. Therefore, based on this premise, a good neural network that meets the actual needs of users can be obtained, which indicates that the improved method based on evolutionary neural network is effective to measure the performance of regional innovation system.     

Research and application of virtual user context information security strategy based on group intelligent computing

This article first introduced the current technology of the privacy protection model, and analyzed their characteristics and deficiencies. Afterwards, from the point of view of revenue, the shortcomings of the traditional privacy protection model have analyzed through the group intelligent computing method. Based on this, this paper proposes a research and application of virtual user information of security strategy based on group intelligent computing, through the collection of visitor's private information historical access data, intelligent calculation of the strategy group between the visitor and the interviewee. The setting of the threshold of the visited person can protect the privacy information of the user more effectively. In this paper, the implementation flow, algorithm implementation process, and specific architecture design of the proposed virtual user of privacy protection model based on group intelligent computing are introduced respectively. The specific algorithms include PCA, BP neural network, and genetic algorithm. Finally, the proposed privacy has verified through experiments. Protection model can protect user privacy more effectively than traditional privacy protection model. In the future, we will further expand and improve the privacy protection model of virtual users based on group intelligent computing, including considering the dynamic and inconsistency of access to the privacy information, that is, accessing different private information will produce different overlay effects and parallelism. We will also study how to apply this model to actual systems such as shopping websites and social platforms, and use commercial data to evaluate the performance of the model and further improve it.