no code implementations • 9 Jul 2023 • Lukasz Korycki, Bartosz Krawczyk
Continual learning models for stationary data focus on learning and retaining concepts coming to them in a sequential manner.
1 code implementation • 15 Dec 2022 • Damien A. Dablain, Colin Bellinger, Bartosz Krawczyk, David W. Aha, Nitesh V. Chawla
We propose a set of techniques that can be used by both deep learning model users to identify, visualize and understand class prototypes, sub-concepts and outlier instances; and by imbalanced learning algorithm developers to detect features and class exemplars that are key to model performance.
1 code implementation • 13 Jul 2022 • Damien Dablain, Colin Bellinger, Bartosz Krawczyk, Nitesh Chawla
We empirically study a convolutional neural network's internal representation of imbalanced image data and measure the generalization gap between a model's feature embeddings in the training and test sets, showing that the gap is wider for minority classes.
1 code implementation • 13 Jul 2022 • Damien Dablain, Bartosz Krawczyk, Nitesh Chawla
A key element in achieving algorithmic fairness with respect to protected groups is the simultaneous reduction of class and protected group imbalance in the underlying training data, which facilitates increases in both model accuracy and fairness.
1 code implementation • 7 Apr 2022 • Gabriel Aguiar, Bartosz Krawczyk, Alberto Cano
This leads to a large-scale experimental study comparing state-of-the-art classifiers in the data stream mining domain.
1 code implementation • 29 Jul 2021 • Kushankur Ghosh, Colin Bellinger, Roberto Corizzo, Bartosz Krawczyk, Nathalie Japkowicz
Structural concept complexity, class overlap, and data scarcity are some of the most important factors influencing the performance of classifiers under class imbalance conditions.
1 code implementation • 24 Jul 2021 • William C. Sleeman IV, Bartosz Krawczyk
This allows us to gain insight into the usefulness of specific components of oversampling algorithms for big data, as well as formulate guidelines and recommendations for designing future resampling approaches for massive imbalanced data.
1 code implementation • 5 May 2021 • Damien Dablain, Bartosz Krawczyk, Nitesh V. Chawla
An important advantage of DeepSMOTE over GAN-based oversampling is that DeepSMOTE does not require a discriminator, and it generates high-quality artificial images that are both information-rich and suitable for visual inspection.
1 code implementation • 24 Apr 2021 • Łukasz Korycki, Bartosz Krawczyk
Data stream mining focuses on adaptation to concept drift and discarding outdated information, assuming that only the most recent data is relevant.
no code implementations • 20 Apr 2021 • Łukasz Korycki, Bartosz Krawczyk
Due to its trainable nature, it is capable of following changes in a stream and evolving class roles, as well as it can deal with local concept drift occurring in minority classes.
1 code implementation • 14 Oct 2020 • Łukasz Korycki, Bartosz Krawczyk
The conducted experiments show that the deep forest approach can be effectively transformed into an online algorithm, forming a model that outperforms all state-of-the-art shallow adaptive classifiers, especially for high-dimensional complex streams.
1 code implementation • 20 Sep 2020 • Łukasz Korycki, Bartosz Krawczyk
Continual learning from streaming data sources becomes more and more popular due to the increasing number of online tools and systems.
no code implementations • 20 Sep 2020 • Łukasz Korycki, Bartosz Krawczyk
In this paper, we propose a framework for robust concept drift detection in the presence of adversarial and poisoning attacks.
no code implementations • 7 Apr 2020 • Michał Koziarski, Michał Woźniak, Bartosz Krawczyk
The proposed method utilizes an energy-based approach to modeling the regions suitable for oversampling, less affected by small disjuncts and outliers than SMOTE.
no code implementations • 17 Nov 2018 • José-Ramón Cano, Pedro Antonio Gutiérrez, Bartosz Krawczyk, Michał Woźniak, Salvador García
Currently, knowledge discovery in databases is an essential step to identify valid, novel and useful patterns for decision making.
no code implementations • 27 Mar 2018 • Anabel Gómez-Ríos, Siham Tabik, Julián Luengo, ASM Shihavuddin, Bartosz Krawczyk, Francisco Herrera
The recognition of coral species based on underwater texture images pose a significant difficulty for machine learning algorithms, due to the three following challenges embedded in the nature of this data: 1) datasets do not include information about the global structure of the coral; 2) several species of coral have very similar characteristics; and 3) defining the spatial borders between classes is difficult as many corals tend to appear together in groups.