The tasks of optimizing asset allocation considering transaction costs can be formulated into the framework of Markov Decision Pro-cesses(MDPs) and reinforcement learning. In this paper, a risk-averse reinforcement le...
详细信息
This paper presents a new distributed data clustering algorithm, which operates successfully on huge data sets. The algorithm is designed based on a classical clustering algorithm, called PAM [8, 9] and a spanning tre...
详细信息
This paper proposes Ε -descending support vector machines (Ε - DSVMs) to model non-stationary financial time series. The Ε -DSVMs are obtained by taking into account the problem domain knowledge of non- stationarit...
详细信息
This paper deals with the application of saliency analysis to Support Vector Machines (SVMs) for feature selection. The importance of feature is ranked by evaluating the sensitivity of the network output to the featur...
详细信息
Caption graphically superimposed in news video frames can provide important indexing information. The automatic extraction and recognition of news captions can be of great help in querying topics of interest in a digi...
详细信息
In applications where preferences are sought it is desirable to order instances of important phenomenon rather than classify them. Here we consider the problem of learning how to order instances based on spatial parti...
详细信息
Backpropagation is often used as the learning algorithm in layered-structure neural networks, because of its efficiency. However, backpropagation is not free from problems. The learning process sometimes gets trapped ...
详细信息
Computer Assisted learning (CAL) successfully merges with many new fields of research. One is the field of artificial intelligence which strongly supports development of intelligent CAL (ICAL). We have developed our a...
详细信息
Computer Assisted learning (CAL) successfully merges with many new fields of research. One is the field of artificial intelligence which strongly supports development of intelligent CAL (ICAL). We have developed our approach to intelligent tutoring implemented as ICAL. The approach can be described as intelligent computerized speaking tutor that supports learning based on experiments with virtual dynamic systems. The approach is suitable for learning the behavior of any dynamic system, especially in the field of complex, live, biomedical systems. We are now implementing this approach in the field of biomedicine. Two important systems are our virtual system GLUCOMAT-homeostatic glucose regulation in human, and ECOLOG-population growth in natural ecosystem.
A market of two-dimensional agents with geographical constraint is modeled with the soap froth analogy and numerical simulations have been performed using a cellular network generated by Voronoi tessellation. By tunin...
详细信息
This paper proposes texture-based text location methods with a neural network (NN) and a Support Vector Machine (SVM). Both a NN and an SVM are employed to train a set of texture discrimination masks for the given tex...
详细信息
暂无评论