Pawel Raif

Motivated learning for the development of autonomous systems

A new machine learning approach known as motivated learning (ML) is presented in this work. Motivated learning drives a machine to develop abstract motivations and choose its own goals. ML also provides a self-organizing system that controls a machines behavior based on competition between dynamically-changing pain signals. This provides an interplay of externally driven and internally generated control signals. It is demonstrated that ML not only yields a more sophisticated learning mechanism and system of values than reinforcement learning (RL), but is also more efficient in learning complex relations and delivers better performance than RL in dynamically-changing environments. In addition, this paper shows the basic neural network structures used to create abstract motivations, higher level goals, and subgoals. Finally, simulation results show comparisons between ML and RL in environments of gradually increasing sophistication and levels of difficulty.

Motivated learning in autonomous systems

Motivated learning (ML) is a new biologically inspired machine learning method. It is the combination of a reinforcement learning (RL) algorithm and a system that creates hierarchy of goals. The goal creation system is concerned with creating new internal goals, building a hierarchy of them, and controlling the agents behavior according to this constituted hierarchy of goals. As in case of reinforcement learning method, a motivated learning agent is learning through interaction with the environment. The comparisons of both methods in special type test environment show that the motivated learning method is more efficient in learning complex relations between available resources (concepts). ML has better performance than RL, especially in dynamically changing environments. In the presented experiments we have shown that ML based agent, which has the ability to set its internal goals autonomously, is able to fulfill the designers goals more effectively than RL based agent. In addition, because the observed concepts are not predefined but emerge during the learning process, this method also addresses problem of merging connectionist and symbolic approaches for intelligent autonomous systems.

Wireless sensor networks in application to patients health monitoring

This work presents an application of Wireless Sensor Network (WSN) of random access with one-way transmission to the monitoring of hospital patients. In the paper, we consider WSN single-hop network using one single radio frequency, such that all nodes are divided into several groups depending on the average time between the transmission due to the different state of health of patients. We apply the Poisson Arrivals See Time Averages (PASTA) to modeling of WSN. We present formula for the probability of collisions which has been verified by simulation studies of network.

Service Oriented Architecture Based Web Application Model for Collaborative Biomedical Signal Analysis

The rapid growth in availability of new biomedical systems and devices capable of acquiring biosignals for disease diagnosis and health monitoring require rigorous processing. Biomedical research by nature depends on integrated problem solving software environment and often involves people located at different geographical positions. The reusability of different personalized tools are limited due to the complex architectural constrains and restricted interoperability among different devices mostly requiring individual tools. Thus, new computational environments are required to provide robust, user friendly, and scalable systems capable of interoperate seamlessly. This work proposes a service oriented architecture (SOA) based web application model for collaborative biosignal analysis and research to facilitate the seamless integration of various existing tools and different Health Information Systems.

Cognitive agent and its implementation in the blender game engine environment

The paper presents a structural model of a cognitive agent and its Blender implementation. Built in a virtual world, the agent is able to act autonomously, observe its environment and learn from its actions using principles of motivated learning. We discuss both its organizational structure and tools we used to develop virtual implementation of the agent. In this paper, we explain why it is important to consider cognitive agents model together with restrictions imposed by its sensory and motor functions and the properties of objects the agent interacts with in the virtual world.

Polymorphic Variants of TNFR2 Gene in Schizophrenia and Its Interaction with -308G/A TNF-α Gene Polymorphism

Aim Many data showed a role of inflammation and dysfunction of immune system as important factors in the risk of schizophrenia. The TNFR2 receptor is a molecule that adapts to both areas. Tumor necrosis factor receptor 2 (TNFR2) is a receptor for the TNF-α cytokine which is a strong candidate gene for schizophrenia. The serum level of TNFR2 was significantly increased in schizophrenia and associated with more severe symptoms of schizophrenia. Methods We examined the association of the three single nucleotide polymorphisms (rs3397, rs1061622, and rs1061624) in TNFR2 gene with a predisposition to and psychopathology of paranoid schizophrenia in Caucasian population. The psychopathology was measured by a five-factor model of the PANSS scale. We also assessed a haplotype analysis with the -308G/A of TNF-α gene. Results Our case-control study (401 patients and 657 controls) revealed that the genetic variants of rs3397, rs1061622, and rs1061624 in the TNFR2 gene are associated with a higher risk of developing schizophrenia and more severe course in men. However, the genotypes with polymorphic allele for rs3397 SNP are protective for women. The rs1061624 SNP might modulate the appearance of the disease in relatives of people with schizophrenia. The CTGG haplotype build with tested SNPs of TNFR2 and SNP -308G/A of TNF-α has an association with a risk of schizophrenia in Caucasian population depending on sex. Our finding is especially true for the paranoid subtypes of schizophrenia.

Association Study of Tumor Necrosis Factor Receptor 1 (TNFR1) Gene Polymorphisms with Schizophrenia in the Polish Population

Schizophrenia is a devastating mental disorder with undetermined aetiology. Previous research has suggested that dysregulation of proinflammatory cytokines and their receptors plays a role in developing schizophrenia. We examined the association of the three single nucleotide polymorphisms (SNPs; rs4149576, rs4149577, and rs1860545) in the tumor necrosis factor receptor 1 (TNFR1) gene with the development and psychopathology of paranoid schizophrenia in the Polish Caucasian sample consisting of 388 patients and 657 control subjects. The psychopathology was assessed using a five-factor model of the Positive and Negative Syndrome Scale (PANSS). SNPs were genotyped using the TaqMan 5′-exonuclease allelic discrimination assay. The SNPs tested were not associated with a predisposition to paranoid schizophrenia in either the entire sample or after stratification according to gender. However, rs4149577 and rs1860545 SNPs were associated with the intensity of the PANSS excitement symptoms in men, which may contribute to the risk of violent behavior. Polymorphisms in the TNFR1 gene may have an impact on the symptomatology of schizophrenia in men.

The Face Tracking System for Rehabilitation Robotics Applications

The paper presents the working model of the face tracking system. The proposed solution may be used as one of the parts of the rehabilitation or assistive robotic system and serve as the robotic vision subsystem or as the module controlling robotic arm. It is a low-cost design, it is based on open source hardware and software components. As a hardware base the Raspberry Pi computer was used. The machine vision software is based on Python programming language and OpenCV computer vision library.

Wireless sensor networks with randomized parameters

In this paper we present a model of single-hop type wireless sensor network with random access and oneway transmission. In the paper, we analyze the WSN single-hop network using one single radio frequency, such that all nodes are divided into several groups depending on the average time between the transmissions. We replaced deterministic number of nodes in groups by random variables. We apply the Poisson Arrivals See Time Averages (PASTA) to modeling the WSN. We present the formula for the collision probability in the new model. The purpose of this approach is to match better the network model with random access to real world applications. In the paper we present an application of Wireless Sensor Network with random access and one-way transmission to monitoring hospital patients, such that all nodes are divided into several groups depending on the average time between the transmission due to the different state of health of patients.

The implementation of wireless sensor networks for environmental monitoring of water facilities

In the paper we present original solution for remote monitoring of water facilities using wireless sensor networks. We present both the mathematical model of the sensor network and the results of computer simulation.