Biologically Inspired Cognitive Architectures最新文献

In this paper a cognitive model is presented for sharing behaviour (retweeting) on Twitter, addressing the underlying cognitive and affective processes. The model explains how the use of emotions in addition to information can cause an amplification in the diffusion of this information. It was designed according to a Network-Oriented Modeling approach based on temporal-causal network models. By mathematical analysis of stationary points it was verified that the implemented network model does what is expected from the design of the model. In addition, the equilibrium equations of the network model were solved algebraically by a symbolic solver and the solutions were shown to relate well to empirically expected outcomes. Validation by parameter tuning was also performed, and also shows a good approximation of empirically expected outcomes.

Communication is a fundamental aspect of the interaction among human beings. In particular, our physical behaviors provide a large part of this communication, because it expresses our emotional internal states. Most of these behaviors are autonomous and reactive, detonated by the assessment of a stimuli perceived in the environment. Within this type of self-responding behaviors are the behaviors of basic emotions. In this paper we propose a conceptual model for the generation of self-responding behaviors for virtual creatures inspired by neuroscientific evidence, focusing on those centered on basic emotions. The conceptual model is implemented as a concurrent and parallel distributed system, which allows virtual creatures to adapt to their environment and generate more credible behavior. The results of this implementation are shown in this article through a case study, in which the execution of the process is observed when the creature interacts with the environment.

Empathy has been defined in the scientific literature as the capacity to relate another’s emotional state and assigned to a broad spectrum of cognitive and behavioral abilities. Advances in neuroscience, psychology and ethology made it possible to refine the defined functions of empathy to reach a working definition and a model of empathy. Recently, cognitive science and artificial intelligence communities made attempts to model empathy in artificial agents, which can provide means to test these models and hypotheses. A computational model of empathy not only would help to advance the technological artifacts to be more socially compatible, but also understand the empathy mechanisms, test theories, and address the ethics and morality problems the Artificial Intelligence (AI) community is facing today. In this paper, we will review the empathy research from various fields, gather the requirements for empathic capacity and construct a model of empathy that is suitable for interactive conversational agents.

Behavior adaptation is an integral aspect for autonomous agents to survive in a world where change is normal. Animals change their foraging routines and socializing habits based on predator risks in their environment. Humans adapt their behavior based on current interests, social norms, stress level, health conditions, upcoming deadlines and various other factors. Artificial agents need to effectively adapt to changes in their environment such that they can quickly adjust their behavior to maintain performance in the changed environment. In this paper, we present a multi-level metacognitive model that allows agents to adapt their behavior in various ways based on the resources available for metacognitive processing. As the agent operates at higher levels of this model, the agent is better equipped to adapt to a wider range of changes. The model has been tested on 2 different applications: (i) a reinforcement learner-based agent trying to navigate and collect rewards in a seasonal grid-world environment and (ii) a convolutional neural network-based agent trying to classify the signals in a radio frequency spectrum world and separate them into known modulations and unknown modulations.

The contemporary era demands a progress with respect to manual work or even semi-machine dependence and the desired procession can be provided by Brain Computer Interface (BCI). As the name suggests, BCI is a bridge between the signals that are generated by the thoughts in our brain and the machine that can implement the produced signals into actions. It is a breakthrough invention in the field of brain-mapping science which can adequately help and aid impaired vision, movement, hearing and any damaged functioning of the body that can be thought of as practically. In this paper, a comprehensive survey on the evolution of BCI with a basic introduction of the functioning of the brain has been provided. A detailed extensive revision on the structure of human brain, BCI and its phases, the mechanisms to extract the signals and the algorithms to put the extracted information to use is provided throughout the course of this study. A comparative study of the phases followed by an extensive discussion of the benchmark techniques has been given. The various bottlenecks have been identified and it has been reasoned why most BCI systems remain as mere prototypes. The ongoing research and progress in the field have been studied and detailed in this review.

The paper proposes a novel cognitive architecture that combines cognitive computing and cognitive agent technologies for performing human-like functionality. The system architecture is known as CIT (Cognitive Information Technology). This design takes advantage of cognitive computing to handle Experiential Information (EI) using audio processing, computer vision, natural language processing, text mining, and data mining techniques. The CIT architecture includes human like cognitive agent functionality comprising attention, learning, memory, action selection, and action to handle human like individual and distributed knowledge bases to create rational decisions. The work shows CIT architecture practical implementation through “CIT framework” developed in C# and python language. For validating the system performance, the paper shows CIT based Object Recognition and Question Answering System. This framework is anticipated to advance the quality of artificial intelligent agent based decision-making using human like perception, comprehend and action skills, reducing real world business errors and assuring the correct, accurate, knowledgeable and well-timed human like decisions.

In this paper, we present a Cognitive Manager for urban traffic control, built using MECA, the Multipurpose Enhanced Cognitive Architecture, a cognitive architecture developed by our research group and implemented in the Java language. The Cognitive Manager controls a set of traffic lights in a junction of roads based on information collected from sensors installed on the many lanes feeding the junction. We tested our Junction Manager in 4 different test topologies using the SUMO traffic simulator, and with different traffic loads. The junction manager seeks to optimize the average waiting times for all the cars crossing the junction, while at the same time being able to provide preference to special cars (police cars or firefighters), called Smart Cars, and equipped with special devices that grant them special treatment during the phase allocation policies provided by the architecture. Simulation results provide evidence for an enhanced behavior while compared to fixed-time policies.

This paper first describes a temporal-causal network model for recognition of emotions shown by others. The model can show both normal functioning and dysfunctioning, such as can be the case with certain types of dementia. More specifically the focus of the paper is on a specific type of therapy that has been incorporated in the model (thus becoming adaptive) to study the effects and potentials of this therapy to improve the dysfunctional behaviour. Simulations have been performed to test the model. A mathematical analysis was done which gave evidence that the model as implemented does what it is meant to do. The model can be applied to obtain a virtual patient model to study the way in which recognition of emotions can deviate for certain types of persons, and what a therapy can contribute to improve the situation.