Learning in Nonstationary Environments: A Survey

By Gregory Ditzler, Manuel Roveri, Cesare Alippi, Robi Polikar. 2015

778 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering, Management Science and Operations Research, Meta & Continual Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Electrical and Electronic Engineering

In Management Science and Operations Research

In Meta & Continual Learning

Citations in the corpus

Continual Lifelong Learning with Neural Networks: A Review (2018, background): “The stability–plasticity dilemma regards the extent to which a systemmust be prone to integrate and adapt to new knowledge and, importantly, how this adaptation process should be compensated by internalmechanisms that stabilize andmodulate neural activity to prevent catastrophic forgetting (Ditzler ”
Lifelong Learning of Spatiotemporal Representations With Dual-Memory Recurrent Self-Organization (2018, background): “The extent to which a system must be flexible for learning novel knowledge and stable for preventing the disruption of consolidated knowledge is known as the stabilityplasticity dilemma, which has been extensively studied for both computational and biological systems (e.g., Grossberg, 1980, 2007; Me”
Continual Learning by Modeling Intra-Class Variation (2022, background): “This problem is closely connected to the stability-plasticity dilemma (Ditzler et al., 2015; Grossberg, 1982; Grossberg et al., 2012) in biological systems, where a system should be plastic enough to absorb new knowledge and at the same time stable enough not to catastrophically forget previous…”
Incremental on-line learning: A review and comparison of state of the art algorithms (2018, background): “Incremental learning is more frequently treated in the setting of streaming scenarios [17, 18], although most of the work particularly targets concept drift [19, 20, 6].”
Continual Learning with Bayesian Neural Networks for Non-Stationary Data (2020, background): “Examples include models of energy demand, climate analysis, ﬁnancial market, or user-behavior analytics (Ditzler et al., 2015).”
A Survey on Ensemble Learning for Data Stream Classification (2017, background): “stream learning and concept drift adaptation [Tsymbal 2004; Žliobaitė 2010; Hoens et al. 2012; Gama et al. 2014; Lemaire et al. 2015; Ditzler et al. 2015; Barddal et al. 2016; Webb et al. 2016].”
Credit Card Fraud Detection: A Realistic Modeling and a Novel Learning Strategy (2018, background): “2) Concept Drift: There are two main factors introducing changes/evolutions in the stream of credit card transactions, which in the literature are typically referred to as concept drift [27], [35].”
Machine Learning at the Network Edge: A Survey (2019, background): “An important area of research is to create novel machine learning algorithms for handling the non-stationary data streams generated by IoT sensors [36].”
Continuous Domain Adaptation with Variational Domain-Agnostic Feature Replay (2020, background): “, 2007), and many others summarized in the survey papers (Gama et al., 2014; Ditzler et al., 2015).”
Continual Learning with Gated Incremental Memories for sequential data processing (2020, background): “Dynamic environments are often subjected to the concept drift phenomenon [1]–[5] which reflects substantial changes in the data generating process and the corresponding predictions.”
Minimax Classification under Concept Drift with Multidimensional Adaptation and Performance Guarantees (2022, neutral): “The statistical characteristics describing the underlying distribution of instance-label pairs often change with time in practical scenarios of supervised classiﬁcation (Gama et al., 2014; Ditzler et al., 2015).”
Mitigating Concept Drift via Rejection (2018, background): “A number of approaches have been proposed for learning in non-stationary environments in the presence of concept drift, whereby several recent technologies are also suited for heterogeneous types of drift [10, 11, 12, 13, 14].”
Analyzing concept drift and shift from sample data (2018, methodology): “We have a growing body of sophisticated methods for learning in the context of concept drift (Gaber et al. 2005; Gama and Rodrigues 2009; Aggarwal 2009; Žliobaite 2010; Bifet et al. 2011; Nguyen et al. 2015; Brzezinski and Stefanowski 2014; Krempl et al. 2014; Gama et al. 2014; Ditzler et al. 2015).”
Unleashing the Power of Edge-Cloud Generative AI in Mobile Networks: A Survey of AIGC Services (2023, neutral): “AIGC models often require frequent fine-tuning and retraining for newly generated data and dynamic requests in nonstationary mobile edge networks [142].”
Class-Incremental Learning via Dual Augmentation (2021, background): “Yet the need of the ability to acquire sequential experience in dynamic and open environments [4, 5, 6] poses a serious challenge to modern deep learning systems, which only perform well on homogenized, balanced, and shufﬂed data [7].”
Ensemble learning for data stream analysis: A survey (2017, background): “The paper [43] postulates that the idea of the so called limited gradual drift is used rather in an intuitive way in most”
From Concept Drift to Model Degradation: An Overview on Performance-Aware Drift Detectors (2022, background): “[15] surveyed the studies on concept drift approaches from two main aspects, active and passive.”
Incremental learning algorithms and applications (2016, methodology): “Changes in the data distribution over time are commonly referred to as concept drift [88, 157, 33, 126].”