Concept drift, i.e., the change of the data generating distribution, can render machine learning models inaccurate. Several works address the phenomenon of concept drift in the streaming context usually assuming that consecutive data points are independent of each other. To generalize to dependent data, many authors link the notion of concept drift to time series. In this work, we show that the temporal dependencies are strongly influencing the sampling process. Thus, the used definitions need major modifications. In particular, we show that the notion of stationarity is not suited for this setup and discuss alternatives. We demonstrate that these alternative formal notions describe the observable learning behavior in numerical experiments.

The world surrounding us is subject to constant change. These changes, frequently described as concept drift, influence many industrial and technical processes. As they can lead to malfunctions and other anomalous behavior, which may be safety-critical in many scenarios, detecting and analyzing concept drift is crucial. In this paper, we provide a literature review focusing on concept drift in unsupervised data streams. While many surveys focus on supervised data streams, so far, there is no work reviewing the unsupervised setting. However, this setting is of particular relevance for monitoring and anomaly detection which are directly applicable to many tasks and challenges in engineering. This survey provides a taxonomy of existing work on drift detection. Besides, it covers the current state of research on drift localization in a systematic way. In addition to providing a systematic literature review, this work provides precise mathematical definitions of the considered problems and contains standardized experiments on parametric artificial datasets allowing for a direct comparison of different strategies for detection and localization. Thereby, the suitability of different schemes can be analyzed systematically and guidelines for their usage in real-world scenarios can be provided. Finally, there is a section on the emerging topic of explaining concept drift.

The world that surrounds us is subject to constant change, which also affects the increasing amount of data collected over time, in social media, sensor networks, IoT devices, etc. Those changes, referred to as concept drift, can be caused by seasonal changes, changing demands of individual customers, aging or failing sensors, and many more. As drift constitutes a major issue in many applications, considerable research is focusing on this setting [4]. Depending on the domain of data and application, different drift scenarios might occur: For example, covariate shift refers to the situation that training and test sets have different marginal distributions [9]. In recent years, a large variety of methods for learning in presence of drift has been proposed [4], whereby a majority of the approaches targets supervised learning scenarios.

We present a modelling framework for the investigation of supervised learning in non-stationary environments. Specifically, we model two example types of learning systems: prototype-based Learning Vector Quantization (LVQ) for classification and shallow, layered neural networks for regression tasks. We investigate so-called student teacher scenarios in which the systems are trained from a stream of high-dimensional, labeled data. Properties of the target task are considered to be non-stationary due to drift processes while the training is performed. Different types of concept drift are studied, which affect the density of example inputs only, the target rule itself, or both. By applying methods from statistical physics, we develop a modelling framework for the mathematical analysis of the training dynamics in non-stationary environments. Our results show that standard LVQ algorithms are already suitable for the training in non-stationary environments to a certain extent. However, the application of weight decay as an explicit mechanism of forgetting does not improve the performance under the considered drift processes. Furthermore, we investigate gradient-based training of layered neural networks with sigmoidal activation functions and compare with the use of rectified linear units (ReLU). Our findings show that the sensitivity to concept drift and the effectiveness of weight decay differs significantly between the two types of activation function.

December 5, 2019 Abstract The notion of drift refers to the phenomenon that the distribution, which is underlying the observed data, changes over time. Albeit many attempts were made to deal with drift, formal notions of drift are application-dependent and formulated in various degrees of abstraction and mathematical coherence. In this contribution, we provide a probability theoretical framework, that allows a formalization of drift in continuous time, which subsumes popular notions of drift. It gives rise to a new characterization of drift in terms of stochastic dependency between data and time. This particularly intuitive formalization enables us to design a new, efficient drift detection method. Further, it induces a technology, to decompose observed data into a drifting and a non-drifting part. Keywords: Online learning, learning theory, stochastic processes, learning with drift, continuous time models, drift decomposition 1 INTRODUCTION One fundamental assumption in classical machine learning is the fact that observed data are i.i.d. Yet, this assumption is often violated as soon as machine learning faces real world problems: models are subject to seasonal changes, changed demands of individual costumers, ageing of sensors, etc. In such settings, lifelong model adaptation rather than classical batch learning is required for optimum performance. Since drift, i.e. the fact that data is no longer identically distributed, is a major issue in many real-world applications of machine learning, many attempts were made to deal with this setting (Ditzler et al., 2015). Depending on the domain of data and application, the presence of drift is modelled in different ways. As an example, covariate shift refers to the situation of training and test set having different marginal distributions (Gretton et al., 2009). Learning for data streams extends this setting to an unlimited (but usually countable) stream of observed data, mostly in supervised learning scenarios (Gama et al., 2014). Learning technologies for such situations often rely on windowing techniques, and adapt the model based on the characteristics of the data in an observed time window. Active methods explicitly detect drift, usually referring to drift of the classification error, and trigger model adaptation this way, while passive methods continuously adjust the model (Ditzler et al., 2015).