Deep neural networks and tabular data: A survey
Heterogeneous tabular data are the most commonly used form of data and are essential for
numerous critical and computationally demanding applications. On homogeneous datasets …
numerous critical and computationally demanding applications. On homogeneous datasets …
Learning parameter distributions to detect concept drift in data streams
Data distributions in streaming environments are usually not stationary. In order to maintain
a high predictive quality at all times, online learning models need to adapt to distributional …
a high predictive quality at all times, online learning models need to adapt to distributional …
On baselines for local feature attributions
High-performing predictive models, such as neural nets, usually operate as black boxes,
which raises serious concerns about their interpretability. Local feature attribution methods …
which raises serious concerns about their interpretability. Local feature attribution methods …
Change detection for local explainability in evolving data streams
As complex machine learning models are increasingly used in sensitive applications like
banking, trading or credit scoring, there is a growing demand for reliable explanation …
banking, trading or credit scoring, there is a growing demand for reliable explanation …
Dynamic model tree for interpretable data stream learning
J Haug, K Broelemann… - 2022 IEEE 38th …, 2022 - ieeexplore.ieee.org
Data streams are ubiquitous in modern business and society. In practice, data streams may
evolve over time and cannot be stored indefinitely. Effective and transparent machine …
evolve over time and cannot be stored indefinitely. Effective and transparent machine …
Standardized Evaluation of Machine Learning Methods for Evolving Data Streams
Due to the unspecified and dynamic nature of data streams, online machine learning
requires powerful and flexible solutions. However, evaluating online machine learning …
requires powerful and flexible solutions. However, evaluating online machine learning …
Online feature screening for data streams with concept drift
Screening feature selection methods are often used as a preprocessing step for reducing
the number of variables before training a model. Traditional screening methods only focus …
the number of variables before training a model. Traditional screening methods only focus …
Employing Two-Dimensional Word Embedding for Difficult Tabular Data Stream Classification
P Zyblewski - arXiv preprint arXiv:2404.15836, 2024 - arxiv.org
Rapid technological advances are inherently linked to the increased amount of data, a
substantial portion of which can be interpreted as data stream, capable of exhibiting the …
substantial portion of which can be interpreted as data stream, capable of exhibiting the …
Towards Reliable Machine Learning in Evolving Data Streams
JC Haug - 2022 - tobias-lib.ub.uni-tuebingen.de
Data streams are ubiquitous in many areas of modern life. For example, applications in
healthcare, education, finance, or advertising often deal with large-scale and evolving data …
healthcare, education, finance, or advertising often deal with large-scale and evolving data …
Drifter: Efficient Online Feature Monitoring for Improved Data Integrity in Large-Scale Recommendation Systems
B Škrlj, N Ki-Tov, L Edelist, N Silberstein… - arXiv preprint arXiv …, 2023 - arxiv.org
Real-world production systems often grapple with maintaining data quality in large-scale,
dynamic streams. We introduce Drifter, an efficient and lightweight system for online feature …
dynamic streams. We introduce Drifter, an efficient and lightweight system for online feature …