Drawing attributions from evolved counterfactuals
JF Jakubik, H Kwaśnicka
Proceedings of the Genetic and Evolutionary Computation Conference Companion
2024
Jan Jakubik
Assistant
Research/education assistant, PhD Student, interested in applications of machine learning to music and audio analysis. Other research interest include methods of learning with insufficient labels, such as semi-supervised and weakly supervised learning. He participated in R&D projects regarding the use of ML to analyse music for videogame content creation and applications of weak supervision in Natural Language Processing of sensitive data.
Research interests
- Deep learnning
- Audio analysis
- Generative media
Research groups
Education
- 2015
MsC in Computer Science
Wroclaw University of Science and Technology
Publications11
Searching for loops and sound samples with feature learning
JF Jakubik
Polskie Towarzystwo Informatyczne 13–18
2022
Retrieving Sound Samples of Subjective Interest With User Interaction
JF Jakubik
Annals of Computer Science and Information Systems · Cited by 2
2020
A neural network approach to Hearthstone win rate prediction
JF Jakubik
Polskie Towarzystwo Informatyczne; Los Alamitos, Ca 185–188 · Cited by 5
2018
Dataset enhancement in hair follicle detection: ESENSEI challenge
JF Jakubik
Annals of Computer Science and Information Systems · Cited by 7
2018
Similarity-based summarization of music files for support vector machines
JF Jakubik, H Kwaśnicka
Complexity · Cited by 5
2018
Evaluation of gated recurrent neural networks in music classification tasks
JF Jakubik
Conference paper · Cited by 11
2018
Evaluation of Hearthstone game states with neural networks and sparse autoencoding
JF Jakubik
Annals of Computer Science and Information Systems · Cited by 4
2017
Music emotion analysis using semantic embedding recurrent neural networks
JF Jakubik, H Kwaśnicka
INISTA 2017 · Cited by 14
2017
Sparse coding methods for music induced emotion recognition
JF Jakubik, H Kwaśnicka
Annals of Computer Science and Information Systems · Cited by 4
2016
Consensus method for integration of knowledge expressed in form of modal formulas
G Skorupa, JF Jakubik, B Fidrysiak
Information systems architecture and technology 43–52
2013