Application of neural networks and machine learning in networkdesign

Communication network design is becoming increasingly complex, involving making networks more usable, affordable, and reliable. To help with this, we have proposed an expert network designer (END) for configuring, modeling, simulating, and evaluating large structured computer networks, employing artificial intelligence, knowledge representation, and network simulation tools. We present a neural network/knowledge acquisition machine-learning approach to improve the END's efficiency in solving the network design problem and to extend its scope to acquire new networking technologies, learn new network design techniques, and update the specifications of existing technologies     

Charting of geoelectric potential signal dynamics via geometrical techniques and its possible relation to significant earthquakes in Western Greece

An exploratory data analysis (EDA) framework was developed in order to investigate the presence of possible correlations in the long-term geoelectric potential (LTGP) signal, prior to significant seismic events in Western Greece. Experimental data, which had been collected continuously within a 6-year period (1998–2003) at the earthquake signal-acquisition center of the University of Patras Seismological Laboratory (UPSL), were studied. The multidimensional scaling (MDS) technique was used to detect putative clustering tendencies in the ensemble of time-series (TS) segments that could indicate the existence of a possible correlation between the geoelectric signal structure and the impending seismic activity. The end result is a 2-dimensional point diagram with the geometry reflecting important signal characteristics, directly related with the underlying seismic events. In addition, the “appending technique”, a well-suited data analysis methodology was incorporated as a means of relating any other TS segment directly with the pre-computed, via the MDS, point diagram. Using the proposed EDA framework, we examined the presence of similar temporal patterns in the recorded geoelectric TS field, 10 days before the occurrence of the major earthquakes. The presented experimental results indicated clearly the validity of the introduced approach and, further, motivated the use of such advanced visualization schemes for charting precursor trends of significance seismicity.     

A NOTE ON INCORPORATING TOOL LIFE VARIABILITY IN THE GEOMETRIC PROGRAMMING FORMULATION OF MACHINING ECONOMICS PROBLEMS

This note incorporates tool life variability in the geometric programming formulation of machining economics problems by transforming the probability of tool failure while machining a part to a function of the cutting variables for both Markovian and non-Markovian tool failure processes. Numerical examples show that the stochastic nature of tool life cannot be ignored, since it results in different optimal cutting variable values compared to the deterministic tool life case.     

Microstructured and Electro-Assisted High-Temperature Wettability of MgO in Contact with a Silicate Slag-Based on Fayalite

In this work, MgO samples have been microstructured by laser treatment and surfaces with a controlled stripe design have been manufactured. Sessile drop wetting measurements were performed in a special heating microscope by determining horizontal contact angles as well as contact angles as a function of surface inclination angles. In a further step, voltage was applied to ground MgO surfaces and the wetting behavior was registered as a function of the voltage level. Activation energy analysis was conducted for all the experiments.     

ON THE CHEMICAL OSCILLATIONS IN DIFFERENTIAL REACTORS: A CASE STUDY OF AMMONIA OXIDATION ON PLATINUM

Trying to identify the mechanism(s) responsible for isothermal oscillatory behavior of heterogeneous catalytic reaction systems, many investigators have assumed low conversions with respect to the limiting reactant(s) so that the bulk phase concentrations of the reactants may be considered constant. In this study, it is shown that even in a differential catalytic CSTR, in which conversion of the reactants may be less than %, the changes of the gas phase concentrations of the reacting.species may play a key role in the overall dynamic behavior of a reaction system; for instance, they may drive the oscillations observed. The reduction of nitric oxide by ammonia on polycrystalline platinum is discussed in detail within this context. Several other oxidation reactions are also discussed     

Rollout sampling approximate policy iteration

Several researchers have recently investigated the connection between reinforcement learning and classification. We are motivated by proposals of approximate policy iteration schemes without value functions, which focus on policy representation using classifiers and address policy learning as a supervised learning problem. This paper proposes variants of an improved policy iteration scheme which addresses the core sampling problem in evaluating a policy through simulation as a multi-armed bandit machine. The resulting algorithm offers comparable performance to the previous algorithm achieved, however, with significantly less computational effort. An order of magnitude improvement is demonstrated experimentally in two standard reinforcement learning domains: inverted pendulum and mountain-car.     

Algorithm,software, and hardware optimizations for Delaunay mesh generation on simultaneous multithreaded architectures

This article focuses on the optimization of PCDM, a parallel, two-dimensional (2D) Delaunay mesh generation application, and its interaction with parallel architectures based on simultaneous multithreading (SMT) processors. We first present the step-by-step effect of a series of optimizations on performance. These optimizations improve the performance of PCDM by up to a factor of six. They target issues that very often limit the performance of scientific computing codes. We then evaluate the interaction of PCDM with a real SMT-based SMP system, using both high-level metrics, such as execution time, and low-level information from hardware performance counters.     

Optimal Control of Two-Stage Discrete Event Systems with Real-Time Constraints

We consider discrete event systems (DES) involving tasks with real-time constraints and seek to control processing times so as to minimize a cost function subject to each task meeting its own constraint. When tasks are processed over a single stage, it has been shown that there are structural properties of the optimal sample path that lead to very efficient solutions of such problems. When tasks are processed over multiple stages and are subject to end-to-end real-time constraints, these properties no longer hold and no obvious extensions are known. We consider a two-stage problem with homogeneous cost functions over all tasks at each stage and derive several new optimality properties. These properties lead to the idea of introducing “virtual” deadlines at the first stage, thus partially decoupling the stages so that the known efficient solutions for single-stage problems can be used. We prove that the solution obtained by an iterative virtual deadline algorithm (VDA) converges to the global optimal solution of the two-stage problem and illustrate the efficiency of the VDA through numerical examples.