Jȩdrzej Kozal

Jȩdrzej Kozal
Wroclaw University of Science and Technology | WUT · Faculty of Information and Communication Technology

Master of Science

About

5
Publications
291
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
1
Citation
Introduction
Skills and Expertise
Machine Learning
Neural Networks and Artificial Intelligence
Pattern Recognition
Supervised Learning
Feature Extraction
Computer Vision
Web Development
C++
Python
Classification
Additional affiliations
December 2019 - October 2021
Samsung
  • Warsaw, Poland
Position
  • Junior Machine Learning Engineer
Education
February 2018 - July 2019
Wroclaw University of Science and Technology
Wroclaw University of Science and Technology
Field of study
  • Computer Science
October 2014 - January 2018
Wrocław University of Science and Technology
Field of study
  • Robotics and Control Engineering

Publications

Publications (5)
Fig. 1. Ensemble diversification policies
Accuracy for each extractor trained and evaluated on one dataset for...
Accuracy for each extractor trained and evaluated on one dataset for...
Accuracy for different ensemble construction methods with text attribute.
+4 Accuracy for different ensemble construction methods with title attribute.
Lifelong Learning Natural Language Processing Approach for Multilingual Data Classification
Preprint
Full-text available
  • May 2022
The abundance of information in digital media, which in today's world is the main source of knowledge about current events for the masses, makes it possible to spread disinformation on a larger scale than ever before. Consequently, there is a need to develop novel fake news detection approaches capable of adapting to changing factual contexts and g...
Figure 1: Chunk-Adaptive Restoration visualization. Red line marks the...
Figure 2: Exemplary accuracy for data stream with abrupt concept. Red...
Figure 3: Impact of chunk size on obtained accuracy.
Figure 4: Accuracy for stream-learn data stream (1).
+2 Figure 5: Accuracy for Usenet dataset (25).
Employing chunk size adaptation to overcome concept drift
Article
Full-text available
  • Mar 2022
Modern analytical systems must process streaming data and correctly respond to data distribution changes. The phenomenon of changes in data distributions is called concept drift , and it may harm the quality of the used models. Additionally, the possibility of concept drift appearance causes that the used algorithms must be ready for the continuous...
Fig. 1. Stacking layers for new tasks. (A) Standard neural network...
Increasing Depth of Neural Networks for Life-long Learning
Preprint
Full-text available
  • Feb 2022
Increasing neural network depth is a well-known method for improving neural network performance. Modern deep architectures contain multiple mechanisms that allow hundreds or even thousands of layers to train. This work is trying to answer if extending neural network depth may be beneficial in a life-long learning setting. In particular, we propose...
Figure 1: Chunk-Adaptive Restoration visualization. Red line marks the...
Figure 2: Exemplary accuracy for data stream with abrupt concept. Red...
Figure 3: Impact of chunk size on obtained accuracy.
Figure 4: Accuracy for stream-learn data stream (1).
+3 Figure 5: Accuracy for Usenet dataset (25).
Employing chunk size adaptation to overcome concept drift
Preprint
Full-text available
  • Oct 2021
Modern analytical systems must be ready to process streaming data and correctly respond to data distribution changes. The phenomenon of changes in data distributions is called concept drift, and it may harm the quality of the used models. Additionally, the possibility of concept drift appearance causes that the used algorithms must be ready for the...
Fig. 1. (Left) the schema of a block with no conv on skip connections...
Table 3 .
Conover post-hoc test results. SwA is short for SMOTE with augmentation.
Imbalance Reduction Techniques Applied to ECG Classification Problem
Chapter
Full-text available
  • Oct 2019
In this work we explored capabilities of improving deep learning models performance by reducing the dataset imbalance. For our experiments a highly imbalanced ECG dataset MIT-BIH was used. Multiple approaches were considered. First we introduced mutliclass UMCE, the ensemble designed to deal with imbalanced datasets. Secondly, we studied the impact...

Projects

Project (1)
Lifelong Machine Learning on Data Stream
Project
Objectives: A. New development in the drift concept area. B. Advanced streaming data research using recurrent systems. C. Lifelong learning on streaming data. This project involves researchers from Faculty of Electrical Engineering and Computer Science (FEI), VŠBTechnical University of Ostrava and the Department of Systems and Computer Networks, Wroclaw University of Science and Technology.