Deep Semisupervised Learning-Based Network Anomaly Detection in Heterogeneous Information Systems

The extensive proliferation of modern information services and ubiquitous digitization of society have raised cybersecurity challenges to new levels. With the massive number of connected devices, opportunities for potential network attacks are nearly unlimited. An additional problem is that many low-cost devices are not equipped with effective security protection so that they are easily hacked and applied within a network of bots (botnet) to perform distributed denial of service (DDoS) attacks. In this paper, we propose a novel intrusion detection system (IDS) based on deep learning that aims to identify suspicious behavior in modern heterogeneous information systems. The proposed approach is based on a deep recurrent autoencoder that learns time series of normal network behavior and detects notable network anomalies. An additional feature of the proposed IDS is that it is trained with an optimized dataset, where the number of features is reduced by 94% without classification accuracy loss. Thus, the proposed IDS remains stable in response to slight system perturbations, which do not represent network anomalies. The proposed approach is evaluated under different simulation scenarios and provides a 99% detection accuracy over known datasets while reducing the training time by an order of magnitude.     

Ambiguity and Communication

The ambiguity of a nondeterministic finite automaton (NFA) N for input size n is the maximal number of accepting computations of N for inputs of size n. For every natural number k we construct a family \((L_{r}^{k}\;|\;r\in \mathbb{N})\) of languages which can be recognized by NFA’s with size k?poly(r) and ambiguity O(n ^k ), but \(L_{r}^{k}\) has only NFA’s with size exponential in r, if ambiguity o(n ^k ) is required. In particular, a hierarchy for polynomial ambiguity is obtained, solving a long standing open problem (Ravikumar and Ibarra, SIAM J. Comput. 19:1263–1282, 1989, Leung, SIAM J. Comput. 27:1073–1082, 1998).     

Elo-based learner modeling for the adaptive practice of facts

We investigate applications of learner modeling in a computerized adaptive system for practicing factual knowledge. We focus on areas where learners have widely varying degrees of prior knowledge. We propose a modular approach to the development of such adaptive practice systems: dissecting the system design into an estimation of prior knowledge, an estimation of current knowledge, and the construction of questions. We provide a detailed discussion of learner models for both estimation steps, including a novel use of the Elo rating system for learner modeling. We implemented the proposed approach in a system for practising geography facts; the system is widely used and allows us to perform evaluation of all three modules. We compare the predictive accuracy of different learner models, discuss insights gained from learner modeling, as well as the impact different variants of the system have on learners’ engagement and learning.     

Acoustic Risk Management

Modern civilization moves ahead with strong economical growth. Using a large number of new technologies and new machines in working procedures brings a proportional growth of noise and vibration. Noise and vibration have become an essential part of active safety measures because they are endangering the health of the employee as well as the public. Preventive solutions are ensured in new legislation and standards too. They facilitate high levels of unification of safety requirements for machines as well as the working environment from the acoustics point of view. © 2012 Wiley Periodicals, Inc.     

Relationship between Blood Vessels and Migration of Neuroblasts in the Olfactory Neurogenic Region of the Rodent Brain

Neural precursors originating in the subventricular zone (SVZ), the largest neurogenic region of the adult brain, migrate several millimeters along a restricted migratory pathway, the rostral migratory stream (RMS), toward the olfactory bulb (OB), where they differentiate into interneurons and integrate into the local neuronal circuits. Migration of SVZ-derived neuroblasts in the adult brain differs in many aspects from that in the embryonic period. Unlike in that period, postnatally-generated neuroblasts in the SVZ are able to divide during migration along the RMS, as well as they migrate independently of radial glia. The homophilic mode of migration, i.e., using each other to move, is typical for neuroblast movement in the RMS. In addition, it has recently been demonstrated that specifically-arranged blood vessels navigate SVZ-derived neuroblasts to the OB and provide signals which promote migration. Here we review the development of vasculature in the presumptive neurogenic region of the rodent brain during the embryonic period as well as the development of the vascular scaffold guiding neuroblast migration in the postnatal period, and the significance of blood vessel reorganization during the early postnatal period for proper migration of RMS neuroblasts in adulthood.     

Fracture characteristics of thermally exposed 9Cr–1Mo steel after tensile and impact testing at room temperature

The paper deals with the fracture analysis of thermally exposed 9Cr–1Mo ferritic steel after tensile and impact testing at room temperature. The temperatures of the thermal expositions were 580, 620, and 650 °C, respectively. The duration of annealing experiments varied from 500 to 5000 h. The influence of thermal expositions on Vickers hardness as well as tensile properties was found to be negligible. On the other hand, remarkable effects of the annealing on room temperature impact toughness were observed. Fracture behaviour of the 9Cr–1Mo steel is strongly affected by the presence of precipitates of secondary phases. Fracture surfaces of tensile samples indicate mainly transgranular dimple fracture mechanism. By contrast, the fracture mode of the samples after impact testing is more complex. It shows both – ductile dimple tearing as well as inter-lath decohesion.     

Microstructure and Properties of Nanocrystalline Copper Strengthened by a Low Amount of Al<Subscript>2</Subscript>O<Subscript>3</Subscript> Nanoparticles

Dispersion-strengthened Cu-Al₂O₃ materials have been studied over recent years to find an optimum processing route to obtain a high strength, thermal-stable copper alloy designed for modern applications in electrical engineering. The study analyses the influence of 1 vol.% of alumina content on strengthening the copper matrix. Microstructure of the Cu-Al₂O₃ composite was studied by x-ray diffraction as well as scanning and transmission electron microscopy. The composite shows a homogeneous, thermal-stable nanostructure up to 900 °C due to dispersed alumina nanoparticles. The particles effectively strengthen crystallite/grain boundaries in processes of powder consolidation and annealing of the compact. In contrast to monolithic Cu, the Cu-1 vol.% Al₂O₃ exhibits more than double strength and hardness. The nanocrystalline matrix and the low amount of alumina particles result in a yield strength of 288 MPa and a ductility of 15% which is a good combination for practical utilization of the material.     

GM3 Ganglioside Linked to Neurofibrillary Pathology in a Transgenic Rat Model for Tauopathy

Glycosphingolipids (GSLs) are amphipathic lipids composed of a sphingoid base and a fatty acyl attached to a saccharide moiety. GSLs play an important role in signal transduction, directing proteins within the membrane, cell recognition, and modulation of cell adhesion. Gangliosides and sulfatides belong to a group of acidic GSLs, and numerous studies report their involvement in neurodevelopment, aging, and neurodegeneration. In this study, we used an approach based on hydrophilic interaction liquid chromatography (HILIC) coupled to high-resolution tandem mass spectrometry (HRMS/MS) to characterize the glycosphingolipid profile in rat brain tissue. Then, we screened characterized lipids aiming to identify changes in glycosphingolipid profiles in the normal aging process and tau pathology. Thorough screening of acidic glycosphingolipids in rat brain tissue revealed 117 ganglioside and 36 sulfatide species. Moreover, we found two ganglioside subclasses that were not previously characterized—GT1b-Ac2 and GQ1b-Ac2. The semi-targeted screening revealed significant changes in the levels of sulfatides and GM1a gangliosides during the aging process. In the transgenic SHR24 rat model for tauopathies, we found elevated levels of GM3 gangliosides which may indicate a higher rate of apoptotic processes.     

AMZ Driverless: The full autonomous racing system

This paper presents the algorithms and system architecture of an autonomous racecar. The introduced vehicle is powered by a software stack designed for robustness, reliability, and extensibility. To autonomously race around a previously unknown track, the proposed solution combines state of the art techniques from different fields of robotics. Specifically, perception, estimation, and control are incorporated into one high‐performance autonomous racecar. This complex robotic system, developed by AMZ Driverless and ETH Zürich, finished first overall at each competition we attended: Formula Student Germany 2017, Formula Student Italy 2018 and Formula Student Germany 2018. We discuss the findings and learnings from these competitions and present an experimental evaluation of each module of our solution.