Search

Concept drifts can occur due to various factors such as changes in the environment or sensor degradation, posing significant challenges to machine learning systems by potentially skewing decision-making processes. Therefore, detecting drifts is essential to maintain the integrity and functionality of these systems. Automatic detectors based on statistical information are usually used to accomplish this task. Optimizing drift detectors through tuning is crucial for effective concept drift analysis. However, relying on user expertise or labeled data for supervised tuning can be challenging. In such cases, the use of unsupervised tuning methods becomes a suitable alternative to adapt detectors to evolving data distributions. To address this gap in the literature an unsupervised tuning method was proposed, leveraging a time series segmentation method. This innovative approach aims to alleviate the reliance on labeled data or user expertise traditionally associated with supervised methods, offering a more adaptive and automated means of tuning detectors. Our results demonstrate that our proposed approach outperforms the default configuration in most evaluated cases. Furthermore, we show that our approach can improve the hyperparameter tuning process when the type of drift is known including supervised methods such as Random Search tuning. By adopting our approach, we can achieve better performance in drift detection and improve the accuracy and reliability of systems that rely on this critical task.

The stream mining paradigm has become increasingly popular due to the vast number of algorithms and methodologies it provides to address the current challenges of Internet of Things (IoT) and modern machine learning systems. Change detection algorithms, which focus on identifying drifts in the data distribution during the operation of a machine learning solution, are a crucial aspect of this paradigm. However, selecting the best change detection method for different types of concept drift can be challenging. This work aimed to provide a benchmark for four drift detection algorithms (EDDM, DDM, HDDMW, and HDDMA) for abrupt, gradual, and incremental drift types. To shed light on the capacity and possible trade-offs involved in selecting a concept drift algorithm, we compare their detection capability, detection time, and detection delay. The experiments were carried out using synthetic datasets, where various attributes, such as stream size, the amount of drifts, and drift duration can be controlled and manipulated on our generator of synthetic stream. Our results show that HDDMW provides the best trade-off among all performance indicators, demonstrating superior consistency in detecting abrupt drifts, but has suboptimal time consumption and a limited ability to detect incremental drifts. However, it outperforms other algorithms in detection delay for both abrupt and gradual drifts with an efficient detection performance and detection time performance. [ABSTRACT FROM AUTHOR]

The stream mining paradigm has become increasingly popular due to the vast number of algorithms and methodologies it provides to address the current challenges of Internet of Things (IoT) and modern machine learning systems. Change detection algorithms, which focus on identifying drifts in the data distribution during the operation of a machine learning solution, are a crucial aspect of this paradigm. However, selecting the best change detection method for different types of concept drift can be challenging. This work aimed to provide a benchmark for four drift detection algorithms (EDDM, DDM, HDDMW, and HDDMA) for abrupt, gradual, and incremental drift types. To shed light on the capacity and possible trade-offs involved in selecting a concept drift algorithm, we compare their detection capability, detection time, and detection delay. The experiments were carried out using synthetic datasets, where various attributes, such as stream size, the amount of drifts, and drift duration can be controlled and manipulated on our generator of synthetic stream. Our results show that HDDMW provides the best trade-off among all performance indicators, demonstrating superior consistency in detecting abrupt drifts, but has suboptimal time consumption and a limited ability to detect incremental drifts. However, it outperforms other algorithms in detection delay for both abrupt and gradual drifts with an efficient detection performance and detection time performance.

The physicochemical calculations of the required amount of the GTT grade goptalum catalyst of ozone decomposition and the pressure drop on the catalyst bed under the operating conditions of the gas systems of particle accelerators were presented. The results of calculations were compared with the experimental data obtained in the gas system of the ATLAS detector unit of the Large Hadron Collider. The obtained physicochemical data make it possible to optimize the parameters of ozone filter destructors and the amount of catalyst required for purification of gas mixtures from ozone in detector units of accelerators. [ABSTRACT FROM AUTHOR]

Concept drift is the problem that the statistical properties of the data generating process change over time. Recently, the Time Decaying Adaptive Prediction (TDAP) algorithm 1 1 Scalable Time-Decaying Adaptive Prediction Algorithm. (Tan et al., 2016). was proposed to address the problem of concept drift. TDAP was designed to account for the effect of drifting concepts by discounting the contribution of previous learning examples using an exponentially decaying factor. The drawback of TDAP is that the rate of its decaying factor is required to be manually tuned. To address this drawback, we propose a new adaptive online algorithm, called Follow-the-Regularized-Leader with Adaptive Decaying Proximal (FTRL-ADP). There are two novelties in our approach. First, we derive a rule to automatically update the decaying rate, based on a rigorous theoretical analysis. Second, we use a concept drift detector to identify major drifts and reset the update rule accordingly. Comparative experiments with 14 datasets and 6 other online algorithms show that FTRL-ADP is most advantageous in noisy environments with real drifts. [ABSTRACT FROM AUTHOR]

Abstract: Thirty years ago planar technology was first successfully applied to the manufacture of silicon detectors and this technique led to detectors with better performance than traditional surface barrier and diffused junction types. Today, these detectors are widely used in nuclear and high-energy physics research and are widely deployed in the industrial market as well. These silicon detectors are used in health physics, X-ray fluorescence and diffraction, medical imaging, space applications, and photon (light) detection. As an industrial company, Canberra was challenged to provide detectors with high reliability along with low leakage current, thin entrance windows, a wide range of thicknesses, low noise, and position sensitivity in both single devices and arrays. A new challenge is the readout of pixilated devices. Electronics integrated into or onto the detector chip is a topic of current interest. [Copyright &y& Elsevier]

The current γ-ray imaging system has the disadvantage of a very high cost due to its need for many pixels, amplifiers and control circuits to obtain high resolution. In order to reduce the cost of γ-ray imaging detectors we propose a single-element detector which is able to obtain two-dimensional γ-ray images, and investigate the feasibility of a one-dimensional position-sensitive detector as a first stage. A CdTe semiconductor detector was employed in the single-element position-sensing detector. Because the mobility-lifetime product for electrons is much higher than that for holes, the movement of holes can be neglected and the CdTe detector can be regarded as an electron drift detector. We made a prototype of a simply structured, one-dimensional position-sensitive detector and evaluated its performance. Consequently, it was confirmed that the position-sensitive detector with a single CdTe element and with sub-millimeter resolution was achieved at low cost. [Copyright &y& Elsevier]

We report results on drift velocity monitoring using MOS charge injectors in silicon drift detectors obtained in beam test conditions. The correction of velocity variations as small as 0.03% caused by temperature variations of the order of 0.04 K allowed to get an average space resolution along all the drift path of 28 μm. Preliminary result demonstrating the possibility to correct for temperature gradients along the anode axis are also presented. [Copyright &y& Elsevier]