Biblio

Reinforcement learning (RL) algorithms have been demonstrated to be very attractive tools to train agents to achieve sequential tasks. However, these algorithms require too many training data to converge to be efficiently applied to physical robots. By using a human teacher, the learning process can be made faster and more robust, but the overall performance heavily depends on the quality and availability of teacher demonstrations or instructions. In particular, when these teaching signals are inadequate, the agent may fail to learn an optimal policy. In this paper, we introduce a trust-based interactive task learning approach. We propose an RL architecture able to learn both from environment rewards and from various sparse teaching signals provided by non-expert teachers, using an actor-critic agent, a human model and a trust model. We evaluate the performance of this architecture on 4 different setups using a maze environment with different simulated teachers and show that the benefits of the trust model.

Problem: Today, many methods of influencing on personnel in the communication process are available to social engineers and information security specialists, but in practice it is difficult to say which method and why it is appropriate to use one. Criteria and indicators of effective communication are not formalized. Purpose: to formalize the concept of effective communication, to offer a tool for combining existing methods and means of communication, to formalize the purpose of communication. Methods: Use of the terminal model of a control system for a non-stochastic communication object. Results. Two examples demonstrating the possibility of using the terminal model of the communication control system, which allows you to connect tools and methods of communication, justify the requirements for the structure and feedback of communication, select the necessary communication algorithms depending on the observed response of the communication object. Practical significance: the results of the research can be used in planning and conducting effective communication in the process of information protection, in business, in private relationships and in other areas of human activity.

This paper presents a human model-based feature extraction method for a video surveillance retrieval system. The proposed method extracts, from a normalized scene, object features such as height, speed, and representative color using a simple human model based on multiple-ellipse. Experimental results show that the proposed system can effectively track moving routes of people such as a missing child, an absconder, and a suspect after events.