Cognitive Agents Research Articles

Cognitive agent based fault tolerance in ubiquitous networks: a machine learning approach

Cognitive agent based fault tolerance in ubiquitous networks: a machine learning approach

From Cognitive Agents to Cognitive Systems: Theoretical, Methodological, and Empirical Developments of van Gelder's (1998) "Dynamical Hypothesis".

Over two decades have passed since the publication of van Gelder's (1998) "dynamical hypothesis." In that paper, van Gelder proposed that cognitive agents were not digital computers-per the representational computational approach-but dynamical systems. The evolution of the dynamical hypothesis was driven by parallel advances in three areas. Theoretically, a deeper understanding of genetics, biology, neuroscience, and cognitive science inspired questions about how systems within each domain dynamically interact and extend their effects across spatiotemporal scales. Methodologically, more sophisticated and domain-general tools allowed researchers to discover, model, and quantify system dynamics, structure, and patterns across multiple scales to generate a more comprehensive system-level understanding of behaviors. Empirically, we can analyze a system's behavior while preserving its natural dynamics, revealing evidence that the reductionist approach leads to an incomplete understanding of the components and the overall system. Researchers have traditionally reduced a complex system into its component processes and assumed that the parts can be recombined to explain the whole. These three advances fundamentally altered our understanding of a "cognitive agent:" How their behaviors are driven by long-range coordination across multiple processes, how the interdependent and nested structure of interacting variables produces behaviors that are greater than the sum of its parts, and how environmental constraints shape adaptive yet stable behavioral patterns.

A Procedural Constructive Learning Mechanism with Deep Reinforcement Learning for Cognitive Agents

Recent advancements in AI and deep learning have created a growing demand for artificial agents capable of performing tasks within increasingly complex environments. To address the challenges associated with continuous learning constraints and knowledge capacity in this context, cognitive architectures inspired by human cognition have gained significance. This study contributes to existing research by introducing a cognitive-attentional system employing a constructive neural network-based learning approach for continuous acquisition of procedural knowledge. We replace an incremental tabular Reinforcement Learning algorithm with a constructive neural network deep reinforcement learning mechanism for continuous sensorimotor knowledge acquisition, thereby enhancing the overall learning capacity. The primary emphasis of this modification centers on optimizing memory utilization and reducing training time. Our study presents a learning strategy that amalgamates deep reinforcement learning with procedural learning, mirroring the incremental learning process observed in human sensorimotor development. This approach is embedded within the CONAIM cognitive-attentional architecture, leveraging the cognitive tools of CST. The proposed learning mechanism allows the model to dynamically create and modify elements in its procedural memory, facilitating the reuse of previously acquired functions and procedures. Additionally, it equips the model with the capability to combine learned elements to effectively adapt to complex scenarios. A constructive neural network was employed, initiating with an initial hidden layer comprising one neuron. However, it possesses the capacity to adapt its internal architecture in response to its performance in procedural and sensorimotor learning tasks, inserting new hidden layers or neurons. Experimentation conducted through simulations involving a humanoid robot demonstrates the successful resolution of tasks that were previously unsolved through incremental knowledge acquisition. Throughout the training phase, the constructive agent achieved a minimum of 40% greater rewards and executed 8% more actions when compared to other agents. In the subsequent testing phase, the constructive agent exhibited a 15% increase in the number of actions performed in contrast to its counterparts.

Implementing Cognitive Semantics of Autoepistemic Membership Statements: The Case of Categories with Prototypes

This article presents a model of an architecture of an artificial cognitive agent that performs the function of generating autoepistemic membership statements used to communicate beliefs about the belonging of an observed external object to a category with a prototype. The meaning of statements is described within the model by means of cognitive semantics. The presented proposal builds upon a pre-existing architecture and a semantic model designed for a simpler case of categories without a prototype. The main conclusion is that it is possible to develop an interactive cognitive agent capable of learning about categories with prototypes and producing autoepistemic membership statements fulfilling requirements of Rosch’s standard version of prototype semantics and satisfying pragmatic and logical rules for generating equivalents of these statements in natural languages. Detailed results include the following: an original proposal for an agent’s architecture, a model of an agent’s strategy of learning categories with a prototype, a scheme for determining the computational complexity of particular implementations of the learning strategy, definitions of cognitive semantics for particular cases of autoepistemic membership statements, and an analytical verification of properties of the proposed cognitive semantics. Finally, this article discusses the directions of further development and potential variants of the proposed architecture.

The KK Principle and the Strong Notion of Knowledge: Hintikka’s Arguments for KK Revisited

In his Knowledge and Belief (1962), Hintikka establishes his system of epistemic logic with the KK (Knowing that One Knows, in symbols, Kp→KKp) principle (KK for short). However, his system of epistemic logic and the KK principle are grounded upon his strong notion of knowledge, which requires that knowledge is infallible, that is, it makes further inquiry pointless, and becomes ‘discussion-stopper’; knowledge implies truth, to wit, cognitive agents will not be mistaken in their knowledge; cognitive agents will be ‘perfect logicians’, i.e. they have infinitive capability of logical inference. Hintikka calls the argument for KK from the strong notion of knowledge as the ‘transcendental argument’ for KK. Obviously, the strong notion of knowledge is far away from our ordinary conception of knowledge; based on such a strong sense of knowledge, epistemic logic has met the problem of logical omnipotence, and is difficult to apply in our ordinary cognitive practice. Moreover, there is no close connection of the KK principle with internalism or externalism.

A Cognitive Robotics Implementation of Global Workspace Theory for Episodic Memory Interaction With Consciousness

Artificial general intelligence revived in recent years after people achieved significant advances in machine learning and deep learning. This leads to the thinking of how real intelligence could be created. Consciousness theories believe that general intelligence is essentially conscious, yet no universal definition is agreed upon. In this work, Global Workspace Theory is implemented and integrated with crucial cognitive components. With the focus on episodic memory and inspiration from the nature of episodic memory in psychology and neuroscience, the episodic memory component is implemented within the Global Workspace framework. In our experiment, the robotic agent operates in a real-world interactive context, forming episodic memory and demonstrating static, temporal, and context memory capabilities during interactions. Consciousness in this work engages in all formation, maintenance, and retrieval processes of episodic memory. The novelty and contributions of this work are 1) this work is implementing episodic memory within the consciousness framework, suggesting the sustainable potential of such an integrated approach to cognitive agents with AGI; 2) Regarding the limited examples in consciousness-based cognitive architectures, this work attempts to contribute to the diversity of perspectives and approaches; 3) Extant episodic memory implementations are suffering from various limitations, while this work summarises some key features for modelling episodic memory within a cognitive architecture; 4) Authors discuss the relationship between episodic memory, consciousness, and general intelligence, proposing the compatibility and relationship between machine consciousness and other AGI research. It is believed that a better alignment between them would further boost the fusion of diverse research for achieving desired cognitive machines.

Predicting Autonomous Driving Behavior through Human Factor Considerations in Safety-Critical Events

This paper investigates the ability of autonomous driving systems to predict outcomes by considering human factors like gender, age, and driving experience, particularly in the context of safety-critical events. The primary objective is to equip autonomous vehicles with the capacity to make plausible deductions, handle conflicting data, and adjust their responses in real-time during safety-critical situations. A foundational dataset, which encompasses various driving scenarios such as lane changes, merging, and navigating complex intersections, is employed to enable vehicles to exhibit appropriate behavior and make sound decisions in critical safety events. The deep learning model incorporates personalized cognitive agents for each driver, considering their distinct preferences, characteristics, and requirements. This personalized approach aims to enhance the safety and efficiency of autonomous driving, contributing to the ongoing development of intelligent transportation systems. The efforts made contribute to advancements in safety, efficiency, and overall performance within autonomous driving systems. To describe the causal relationship between external factors like weather conditions and human factors, and safety-critical driver behaviors, various data mining techniques can be applied. One commonly used method is regression analysis. Additionally, correlation analysis is employed to reveal relationships between different factors, helping to identify the strength and direction of their impact on safety-critical driver behavior.

Harnessing Cognitively Inspired Predictive Models to Improve Investment Decision-Making

In the last years, researchers and practitioners have focused on defining portfolio optimization approaches. This task aims to identify a suitable distribution of assets for maximizing profits and minimizing risks, also offering protection against unexpected market behaviors. Nevertheless, the state-of-the-art approaches encounter significant limitations due to the complex nature of the task: (1) forecasting of non-stationary, non-linearity and volatile stock price; (2) budget allocation over different stocks satisfying multi-objective objective function; (3) risk costs can significantly affect the effectiveness of the designed approaches. In this paper, we propose a cognitively inspired framework for portfolio optimization by integrating deep learning-based stock forecasting for maximizing the revenue and portfolio diversification and Shape Ratio for minimizing the risk. Furthermore, the cognitively inspired forecasting module relies on the LSTM-based approach which combines historical financial data and technical indicators. Hence, this approach addresses the portfolio optimization task with the aim of designing more and more cognitive agents that perform autonomous actions for supporting decision-making. To make these agents cognitive, we further integrate stock forecasting into the portfolio optimization model, also investigating the main factors affecting both stock forecasting and portfolio optimization tasks. The proposed framework has been evaluated in two stages on a real-world dataset, composed of four years of information about stocks from six different areas. Firstly, we compare the proposed forecasting models based on LSTM and GRU, pointing out that the former achieves higher effectiveness results although the latter has a shorter training time. Finally, the proposed framework has been compared with different baselines, obtaining a net difference of $168 at the maximum. Finally, we compare the proposed approach w.r.t. several baselines in terms of total revenue, also providing an ablation analysis to investigate how stock prediction might support investors in dealing with portfolio optimization task.

Growing an Embodied Generative Cognitive Agent

An evolutionary perspective on embodiment puts maintenance of physiology within a functional envelope as the brain’s base goal, with all other goals as refinements. Thus, all goals have physiological perturbation for their motivation and allostatic recovery as their signal of fulfillment. From this account, two entailments emerge. First, an object’s properties are not intrinsic to the object but a situated function of the morphology of the object and the affordances required by the goal. Second, categories do not exist without reference to some goal; they are constructed at the time of perception by blending prior conceptual knowledge to create an understanding of the perception with respect to the goal. Our thesis is that generative large language model (LLM) architectures are part of the solution to creating artificial organic-like cognitive architectures, but that LLMs as currently trained are generative only at a surface-level of behavior rather than deeper levels of cognition and, furthermore, that generative architectures must be coupled with an embodied cognitive agent architecture, which suggests both the additional levels at which generativity must operate and capabilities that the combined architecture must support.

Skill Development through Artificial Cognitive Systems and Social Robotics Applied at Tech-Education

The focus of this research is to demonstrate how a platform composed of systems of artificial cognitive agents and social robotics can interact, teach and learn with students and teachers, through a pedagogical practice and methodological integration of the psychological concepts of the Theory of Multiple Intelligences, the educational foundations of the Dialectic methodology and the relationship explored in the Man-Machine Symbiosis. The objective of this article is to present the cognitive methodological concept of the project developed, grounding it from the theoretical principles, to the selection criteria for the models presented in the current discussions, taking as lines of thought that seek to define the strategy for creating personalized learning storytelling, the techniques for building social robotics in education and the interactivity between visual feedback in the challenging context of technological education.

Exploiting Language Models as a Source of Knowledge for Cognitive Agents

Large language models (LLMs) provide capabilities far beyond sentence completion, including question answering, summarization, and natural-language inference. While many of these capabilities have potential application to cognitive systems, our research is exploiting language models as a source of task knowledge for cognitive agents, that is, agents realized via a cognitive architecture. We identify challenges and opportunities for using language models as an external knowledge source for cognitive systems and possible ways to improve the effectiveness of knowledge extraction by integrating extraction with cognitive architecture capabilities, highlighting with examples from our recent work in this area.

Proposed Uses of Generative AI in a Cybersecurity-Focused Soar Agent

With the rapidly increasing use of AI and machine learning in recent years and the current generative AI revolution, it is no surprise that the malicious use of AI has begun to establish it- self in the realm of cybersecurity. At risk of being left behind in this “arms race”, it’s imperative that autonomous intelli- gent cybersecurity agent (AICAs) are developed to counter this emerging threat. Currently, a project at Argonne National Laboratory is using Soar as a starting point for developing a cognitive-architecture based AICA, but the utilization of Soar in this project has shortcomings, in particular the lack of modern AI principles to generate novel analysis in the face of novel situations. Generative AI has the potential to allow a Soar cognitive agent to consider a much broader range of con- textual information, and learn from past episodic knowledge in novel ways by using transformer architectures. This paper focuses on the theoretical integration of Generative AI into the Soar cognitive architecture from a cybersecurity stand- point, and discuss the advantages to doing so.

Flood Risk and Preventive Choices: A Framework for Studying Human Behaviors.

The topic of flood phenomena has always been of considerable importance due to the high risks it entails, both in terms of potential economic and social damage and the jeopardizing of human lives themselves. The spread of climate change is making this topic even more relevant. This work aims to contribute to evaluating the role that human factors can play in responding to critical hydrogeological phenomena. In particular, we introduce an agent-based platform for analyzing social behaviors in these critical situations. In our experiments, we simulate a population that is faced with the risk of a potentially catastrophic event. In this scenario, citizens (modeled through cognitive agents) must assess the risk they face by relying on their sources of information and mutual trust, enabling them to respond effectively. Specifically, our contributions include (1) an analysis of some behavioral profiles of citizens and authorities; (2) the identification of the "dissonance between evaluation and action" effect, wherein an individual may behave differently from what their information sources suggest, despite having full trust in them in situations of particular risk; (3) the possibility of using the social structure as a "social risk absorber", enabling support for a higher level of risk. While the results obtained at this level of abstraction are not exhaustive, they identify phenomena that can occur in real-world scenarios and can be useful in defining general guidelines.

Harnessing sensor fusion and AI for accurate accident detection and classification in the safety of smart cities

With the growing number of automobiles, traffic accidents are increasing daily. The World Health Organization (WHO) study reports that annually, 1.4 million individuals have died, and 50 million have been wounded globally. An advanced accident detection technique using cognitive agents will reduce rescue operational delays, perhaps saving several lives. Intelligent Transportation Systems (ITS) are gaining significant attention in academia and industry because of the increasing popularity of smart cities. They are seen to enhance road safety in these urban areas. Internet of Things (IoT) and Artificial Intelligence (AI) systems have been widely used to decrease the time needed for rescue operations after an accident. This study introduces an IoT-enabled Automotive Accident Detecting and Categorization (IoT-AADC) method that combines a smartphone's internal and external sensors to identify and categorize the kind of accident. This innovative method enhances the effectiveness of emergency support like fire departments, towing agencies, etc., by providing crucial information regarding the accident category for better planning and execution of rescuing and relief activities. Emergency support providers enhance their preparedness by assessing the injuries experienced by those injured and the damage to the automobiles.

Usefulness of Information for Achieving Goals with Disjunctive Premises

Abstract We propose a logic-based framework to model a system whose aim is to help provide the user with those pieces of information that are useful with respect to his/her current information need, as well as relevant to his/her query. More precisely, we propose three measures of information usefulness which take into account the fact that the user can be represented as a cognitive agent endowed with some beliefs—a partial “picture” about what it already knows—and goals—a certain state of affairs in which the agent would like to be. This paper extends a previous version of the framework by considering a more realistic hypothesis, according to which there are several ways to achieve goals. We present three different approaches: the binary approach, the ordinal approach, and the numerical approach. We take information retrieval (IR) as a particular application domain, and we compare some existing measures with the usefulness measure we introduce here.

Role of Islamic Education Teachers in Shaping Students' Religious Character in the Digital Era: A Case Study of SDN 1 Kondangsari, Cirebon

This qualitative study explores the multifaceted role of Islamic Education (PAI) teachers in shaping students' religious character amid the digital era's challenges at SDN 1 Kondangsari, Cirebon. Employing a case study approach within a Field Research design, the research integrates semi-structured interviews, non-participant observations, and document analysis, analyzed through Miles and Huberman's interactive model to ensure a comprehensive understanding. The findings reveal that PAI teachers are facilitators, mentors, environmental providers, communicators, role models, evaluators, innovators, moral and political agents, cognitive agents, and managers in cultivating students' resilient, religious character. They adeptly utilize digital tools, instill Islamic practices through habituation, and embody exemplary conduct, equipping students to navigate modern society's intricacies. The study underscores the importance of integrating traditional religious teachings with digital advancements and calls for further research to assess these pedagogical strategies' broader applicability and long-term impact. The research highlights the critical need for continuous innovation in teaching approaches to bridge the gap between faith and technology, contributing to the field of religious education and societal development. The implications of this study underscore the necessity for comprehensive teacher training programs, supportive school environments, and collaborative efforts involving educators, parents, and the community in character-building initiatives.

Individuals, Collectives, and Individuals in Collectives: The Ineliminable Role of Dependence

Our beliefs are inextricably shaped through communication with others. Furthermore, even conversation we conduct in pairs may itself be taking place across a wider, connected social network. Our communications, and with that our thoughts, are consequently typically those of individuals in collectives. This has fundamental consequences with respect to how our beliefs are shaped. This article examines the role of dependence on our beliefs and seeks to demonstrate its importance with respect to key phenomena involving collectives that have been taken to indicate irrationality. It is argued that (with the benefit of hindsight) these phenomena no longer seem surprising when one considers the multiple dependencies that govern information acquisition and the evaluation of cognitive agents in their normal (i.e., social) context.

A Novel Artificial Social Cognitive Agent Model for Teaching the Concepts

Nowadays, extracting the most suitable information becomes a very challenging problem as accumulation of data over Internet is exponentially increasing. This raises the need of such systems which can process data meaningfully, have capability of understanding the user’s questions, and can provide the most relevant information. The existing systems are not capable of giving precise results. The performance of such systems can be improved using Natural Language Processing (NLP) and Cognitive Computing System. Therefore, an artificial cognitive agent is proposed in this paper using NLP, which can take decisions in real-life using social cognition. These agents are expert in their respective domains. Moreover, social behavioural aspects and the constraints of proposed cognitive agent are also modelled in this work. The performance of the proposed agent is evaluated on the concepts of Electric Motor domain in terms of various parameters. The results depict that the proposed artificial cognitive agent obtains satisfactory results.

Understanding Collective Human Behavior in Social Media Networks Via the Dynamical Hypothesis: Applications to Radicalization and Conspiratorial Beliefs.

The dynamical hypothesis has served to explore the ways in which cognitive agents can be understood dynamically and considered dynamical systems. Originally used to explain simple physical systems as a metaphor for cognition (i.e., the Watt governor) and eventually more complex animal systems (e.g., bird flocks), we argue that the dynamical hypothesis is among the most viable approaches to understanding pressing modern-day issues that arise from collective human behavior in online social networks. First, we discuss how the dynamical hypothesis is positioned to describe, predict, and explain the time-evolving nature of complex systems. Next, we adopt an interdisciplinary perspective to describe how online social networks are appropriately understood as dynamical systems. We introduce a dynamical modeling approach to reveal information about emergent properties in social media, where radicalized conspiratorial beliefs arise via coordination between user-level and community-level comments. Lastly, we contrast how the dynamical hypothesis differs from alternatives in explaining collective human behavior in social networks.

Intelligence Offloading and the Neurosimulation of Developmental Agents

Cognitive offloading occurs when environmental affordances expand cognitive capacity while facilitating spatial and social behaviors. Capacity-related constraints are also important, particularly as embodied agents come online during development. Vast differences in brain size and offloading capacity exist across the tree of life. We take from multiple perspectives to understand the proportional contributions of internal models (brain) and externalized processing (offloading) in developing embodied computational agents. As developing nervous systems scale with body size and/or functional importance, offloading is also driven by neural capacity. Cognitive capacity is ultimately determined by various innate and environmental constraints. We propose a similar model for computationally developing cognitive agents. A regulatory model of cognition is proposed as a means to build cognitive systems that interface with biologically-inspired substrates. Multiple tradeoffs result from energetic, innate, and informational constraints, and determine the proportion of internal to external information processing capacity. As growth of a biologically-inspired substrate accelerates or decelerates over developmental time, it changes the acquisitional capacity of the agent. Our agent’s capacity limitations determine externalization potential, which is characterized by three parameters and two mathematical functions. The neurosimulation approach to intelligence offloading can be applied to a broad range of agent-based models and Artificial Intelligences.