Learning in the presence of concept drift and hidden contexts
Gerhard Widmer,Miroslav Kubat +1 more
TLDR
A family of learning algorithms that flexibly react to concept drift and can take advantage of situations where contexts reappear are described, including a heuristic that constantly monitors the system's behavior.Citations
More filters
Journal ArticleDOI
Implikationen von Machine Learning auf das Datenmanagement in Unternehmen
René Kessler,Jorge Marx Gómez +1 more
TL;DR: There is currently no fully established standard process for the machine learning life cycle, as is the case in data mining with the CRISP-DM-Process, which means that the operationalization of machine learning models in particular can present companies with major challenges.
Book ChapterDOI
Adaptive classifiers with ICI-based adaptive knowledge base management
TL;DR: Experimental results show that the proposed adaptive classification system is particularly effective in situations where the process generating the data evolves through a sequence of abrupt changes.
Proceedings ArticleDOI
Traffic4cast at NeurIPS 2021 - Temporal and Spatial Few-Shot Transfer Learning in Gridded Geo-Spatial Processes
Christian Eichenberger,Moritz Neun,Henry Martin,Pedro Herruzo,Markus Spanring,Yichao Lu,Sung-Ja Choi,V. E. Konyakhin,N.N. Lukashina,Aleksei Shpilman,Nina Wiedemann,Martin Raubal,Bo Wang,Hai Linh Vu,Reza Mohajerpoor,Cheng-Lin Cai,Inhi Kim,L. Hermes,Andrew Melnik,Riza Velioglu,Markus Vieth,Malte Schilling,Alabi Bojesomo,Hasan Al Marzouqi,Panos Liatsis,Jay Santokhi,Dylan Hillier,Yiming Yang,Joned Sarwar,Anna Jordan,Emil Hewage,David Jonietz,Fei Tang,Aleksandra Gruca,Michael Kopp,David P. Kreil,Sepp Hochreiter +36 more
TL;DR: The IARAI Traffic4cast 2021 competition now covers ten cities over 2 years, providing data compiled from over 10 12 GPS probe data, and U-Nets proved to be the winning architecture, demonstrating an ability to extract relevant features in this complex real-world geo-spatial process.
Book ChapterDOI
Robustness of classifiers to changing environments
TL;DR: K-Nearest Neighbor and Artificial Neural Networks are the most robust learners, ensemble algorithms are somewhat robust, whereas Naive Bayes, Logistic Regression and particularly Decision Trees are themost affected.
Journal ArticleDOI
Temporal contexts: Effective text classification in evolving document collections
Leonardo Rocha,Fernando Mourão,Hilton de Oliveira Mota,Thiago Salles,Marcos André Gonçalves,Wagner Meira +5 more
TL;DR: This work highlights the proposal of a pragmatical methodology for evaluating the temporal evolution in ADC domains and presents a strategy, named temporal context selection, for selecting portions of the training set that minimize factors associated to ADC models degradation over time.
References
More filters
Proceedings ArticleDOI
A theory of the learnable
TL;DR: This paper regards learning as the phenomenon of knowledge acquisition in the absence of explicit programming, and gives a precise methodology for studying this phenomenon from a computational viewpoint.
Journal ArticleDOI
Instance-Based Learning Algorithms
TL;DR: This paper describes how storage requirements can be significantly reduced with, at most, minor sacrifices in learning rate and classification accuracy and extends the nearest neighbor algorithm, which has large storage requirements.
Book
Machine Learning: An Artificial Intelligence Approach
TL;DR: This book contains tutorial overviews and research papers on contemporary trends in the area of machine learning viewed from an AI perspective, including learning from examples, modeling human learning strategies, knowledge acquisition for expert systems, learning heuristics, discovery systems, and conceptual data analysis.
Journal ArticleDOI
Learnability and the Vapnik-Chervonenkis dimension
TL;DR: This paper shows that the essential condition for distribution-free learnability is finiteness of the Vapnik-Chervonenkis dimension, a simple combinatorial parameter of the class of concepts to be learned.
Journal ArticleDOI
Queries and Concept Learning
TL;DR: This work considers the problem of using queries to learn an unknown concept, and several types of queries are described and studied: membership, equivalence, subset, superset, disjointness, and exhaustiveness queries.