Dynamic classification approach using scalable ensemble of autoencoders to classify data with drift

The problem of classification under concept drift conditions is investigated. The importance of anomaly detection is emphasized as a key feature of successful approach to operate with adversarial attacks and data poisoning. An approach to classification in the context of both drift and anomalies is introduced. It is based on ensemble of one-class classifiers, implemented by neural network autoencoders. Numeric parameters and supplementary logic are also supposed to distinguish between different classification cases. The quality of classifiers is estimated by original characteristics (EDCA), which examine both training set area and the area around it. The proposed approach is evaluated on synthetic data to highlight its properties in various conditions including normal, drift, new class and anomaly cases.

DEFAULTABLE TERM STRUCTURES DRIVEN BY SEMIMARTINGALES

In this paper, we consider a market with a term structure of credit risky bonds in the single-name case. We aim at minimal assumptions extending existing results in this direction: first, the random field of forward rates is driven by a general semimartingale. Second, the Heath–Jarrow–Morton (HJM) approach is extended with an additional component capturing those future jumps in the term structure which are visible from the current time. Third, the associated recovery scheme is as general as possible, it is only assumed to be nonincreasing. In this general setting, we derive generalized drift conditions which characterize when a given measure is a local martingale measure, thus yielding no asymptotic free lunch with vanishing risk (NAFLVR), the right notion for this large financial market to be free of arbitrage.

Robust Control with Uncertain Disturbances for Vehicle Drift Motions

Professor drivers, including racing drivers, can drive cars to achieve drift motions by taking control of the steering angle in high tire slip ratios, which provides a way to improve the driving safety of autonomous vehicles. The existing studies can be divided into two kinds based on analysis methods, and the theory-based is chosen in this study. Because the recent theory based is most applied for planar models with neglect of the rollover accident risk, the nonlinear vehicle model is established by considering longitudinal, lateral, roll, and yaw motions and rolling safety with the nonlinear tire model UniTire. The drift motion mechanism is analyzed in steady and transient states to obtain drift motion conditions, including the velocity limitation and the relationship between sideslip angle and yaw rate, and vehicle main status parameters including the velocity, side-slip angle and yaw rate in drift conditions. The state-feedback controller is designed based on robust theory and LMI (linear matrix inequation) with uncertain disturbances to realize circle motions in drift conditions. The designed controller in simulations realizes drift circle motions aiming at analyzed status target values by matching the front-wheel steering angle with saturated tire forces, which satisfies the Lyapunov stability with robustness. Robust control in drift conditions solves the problem of how to control vehicles to perform drift motions with uncertain disturbances and improves the driving safety of autonomous vehicles.

Atmospheric drivers of a 4-month drift of an ice buoy in the Antarctic marginal ice zone

<p>Sea-ice drift in the Antarctic marginal ice zone (MIZ) was investigated by using an ice buoy (buoy U1), deployed during the winter sea-ice expansion in July 2017, and drifted for approximately four months from the South Atlantic sector to the Indian Ocean sector of the Southern Ocean. The analysis of this buoy revealed that it remained within the MIZ even during the winter ice expansion, as the mixed pancake-frazil field was maintained. This allowed for a continued assumption of free drift conditions for buoy U1’s full drift, where it continued to respond linearly to the momentum transfer from surface winds. The analysis of buoy U1 also indicated a strong inertial signature at a period of 13.47 hours however, the wavelet analysis indicated majority of the power remained within the lower frequencies. This strong influence at the lower (multi-day) frequencies has therefore been identified as the primary effect of atmospheric forcing. When these lower frequencies were filtered out using the Butterworth high-pass filter it allowed the inertial oscillations to become more significant within the wavelet power spectrum, where it can be seen that these inertial oscillations were often triggered by the passage of cyclones. The initiation of inertial oscillations of sea ice has therefore been identified as the secondary effect of atmospheric forcing, which dominates ice drift at sub-daily timescales and results in the deviation of ice drift from a straight-line path. This comprehensive analysis suggests that the general concentration-based definition of the MIZ is not enough to describe the sea-ice cover, and that the MIZ, where sea ice is in free drift and under the influence of cyclone induced inertial motion, and presumably waves, extends up to »200 km.</p>

Reconstructing the pre–Variscan puzzle of Cambro–Ordovician basement rocks in the western Mediterranean region of Gondwana

<p>In today’s western Mediterranean region, Variscan and Alpine thrusts and shear zones combine to hamper a correct identification and palinspastic reconstruction of Cambro-Ordovician sequences. However, gap-related stratigraphic, climatically sensitive facies associations, sedimentary, volcanosedimentary, biogeographic, biodiversity and detrital zircon data mainly made available during the last two decades allow envisaging a new palaeogeographic scenario by linking proximal-to-distal transects across the western and eastern branches of the Ibero-Armorican Arc. Variscan parautochthonous and autochthonous domains are represented palaeogeographically by, from SW to NE: (i) the Central Iberian, West Asturian-Leonese and Cantabrian zones of the Iberian Massif and their laterally correlative Central Armorican Domain, fringed marginally by the Ossa-Morena and North Armorican thinned outer margin of Gondwana, reminiscent of the rift axis during the Cambrian; and (ii) the southeastern Pyrenees, Occitan and SW Sardinia domains, fringed marginally by the slope-to-basinal South Armorican, Thiviers-Payzac, Albigeois and northeastern Pyrenees domains. These proximal-to distal transects of West Gondwana record a diachronous SW-to-NE migration of evaporites, phosphorites and maximum peak of trilobite diversity, related to the counter-clockwise migration of the Gondwana supercontinent, supported by a gradual modification of detrital zircon provenance. Both branches of the Ibero-Armorican Arc also display a diachronous migration of Cambro-Ordovician rift-to-drift conditions associated with distinct igneous manifestations (volcanosedimentary and plutonic). This migration is related to the development of the Furongian (Toledanian) to Mid-Late Ordovician (Sardic) geodynamic events, in response to gap-related thermal doming, subaerial denudation and magmatic activity evolving from calc-alkaline to tholeiitic affinity.</p>

Evaluation of Automated, In-Cockpit Swath Displacement

HighlightsAutomated swath displacement works well in high drift conditions.System relevance and performance are limited in low drift conditions.Testing against pilot skill without an automated system is necessary.ABSTRACT. A set of spray trials was designed to evaluate the onboard swath displacement feature that is offered as part of some avionic systems and used to aid in targeting aerial spray applications. The trials were flown in Miramichi, New Brunswick, Canada, and were intended to provide basic information on the capabilities of these systems. A total of 32 trials were run, of which 24 provided coverage data that could be used in this evaluation. Two aircraft types were tested, each with distinct avionics systems and automated swath displacement capabilities. Each aircraft was flown with two application setups, resulting in four application scenarios. The systems were generally able to allow the pilot to apply the spray with a swath peak within 5 m of the target line on average, even in high wind conditions. Other relationships that were anticipated in the data, such as a direct correlation between offset distance and targeting accuracy, were not observed. The systems do not appear to add much capability in low drift conditions when not much displacement is expected. It is recognized that a control data set is necessary to evaluate the extent to which the systems improve targeting accuracy. Keywords: Aerial management system, Aerial pesticide application, Drift offset, Precision application, Real-time modeling, Spray deposition, Spray drift, Swath displacement.

Default Ambiguity

This paper discusses ambiguity in the context of single-name credit risk. We focus on uncertainty in the default intensity but also discuss uncertainty in the recovery in a fractional recovery of the market value. This approach is a first step towards integrating uncertainty in credit-risky term structure models and can profit from its simplicity. We derive drift conditions in a Heath–Jarrow–Morton forward rate setting in the case of ambiguous default intensity in combination with zero recovery, and in the case of ambiguous fractional recovery of the market value.