I am a Machine Learning MSc student at the University of Tübingen, working in the STAI group of Seong Joon Oh.

Interests: I am interested in probabilistic model architectures capable of representing different sources of uncertainty. My goal is to contribute to the theoretical foundations of uncertainty in machine learning while developing scalable practical solutions. I am also excited about computer vision.

Bio: I received my BSc degree in Computer Science from ELTE Eötvös Loránd University in 2021 (Grade: Outstanding) with the Best Thesis and Outstanding Student of the Faculty awards. I previously worked on neural program synthesis with a focus on provable correctness. I am currently writing my master's thesis about uncertainty quantification under the supervision of Seong Joon Oh and Michael Kirchhof.

For any inquiries, feel free to reach out to me via mail!

Profile picture

Publications

Project image
Trustworthy Machine Learning
Bálint Mucsányi, Michael Kirchhof, Elisa Nguyen, Alexander Rubinstein, Seong Joon Oh
arXiv.org, 2023
Project Page / Paper /
@InProceedings{Mucsanyi2023ARXIV, 
	author = {Bálint Mucsányi and Michael Kirchhof and Elisa Nguyen and Alexander Rubinstein and Seong Joon Oh}, 
	title = {Trustworthy Machine Learning}, 
	booktitle = {arXiv.org}, 
	year = {2023}, 
}
Project image
URL: A Representation Learning Benchmark for Transferable Uncertainty Estimates
Michael Kirchhof, Bálint Mucsányi, Seong Joon Oh, Enkelejda Kasneci
Advances in Neural Information Processing Systems: Datasets and Benchmarks Track (NeurIPS D&B), 2023
Paper / Code /
@InProceedings{Kirchhof2023NEURIPSDB, 
	author = {Michael Kirchhof and Bálint Mucsányi and Seong Joon Oh and Enkelejda Kasneci}, 
	title = {URL: A Representation Learning Benchmark for Transferable Uncertainty Estimates}, 
	booktitle = {Advances in Neural Information Processing Systems: Datasets and Benchmarks Track (NeurIPS D&B)}, 
	year = {2023}, 
}
Project image
URL: A Representation Learning Benchmark for Transferable Uncertainty Estimates (Best Student Paper Award)
Michael Kirchhof, Bálint Mucsányi, Seong Joon Oh, Enkelejda Kasneci
Uncertainty in Artificial Intelligence Workshop on Epistemic Uncertainty in Artificial Intelligence (E-pi UAI), 2023
Paper / Code /
@InProceedings{Kirchhof2023UAIEAI, 
	author = {Michael Kirchhof and Bálint Mucsányi and Seong Joon Oh and Enkelejda Kasneci}, 
	title = {URL: A Representation Learning Benchmark for Transferable Uncertainty Estimates}, 
	booktitle = {Uncertainty in Artificial Intelligence Workshop on Epistemic Uncertainty in Artificial Intelligence (E-pi UAI)}, 
	year = {2023}, 
}
Project image
Flexible Example-Based Program Synthesis on Tree-Structured Function Compositions
Bálint Mucsányi, Bálint Gyarmathy, Ádám Czapp, Balázs Pintér
Springer Nature Computer Science (SNCS), 2022
Paper / Code /
@InProceedings{Mucsanyi2022SNCS, 
	author = {Bálint Mucsányi and Bálint Gyarmathy and Ádám Czapp and Balázs Pintér}, 
	title = {Flexible Example-Based Program Synthesis on Tree-Structured Function Compositions}, 
	booktitle = {Springer Nature Computer Science (SNCS)}, 
	year = {2022}, 
}
Project image
Flexcoder: Practical Program Synthesis with Flexible Input Lengths and Expressive Lambda Functions (Oral Talk, Best Student Paper Award Finalist)
Bálint Gyarmathy, Bálint Mucsányi, Ádám Czapp, Dávid Szilágyi, Balázs Pintér
arXiv.org, 2021
Paper / Supplemental /
@InProceedings{Gyarmathy2021ICPRAM, 
	author = {Bálint Gyarmathy and Bálint Mucsányi and Ádám Czapp and Dávid Szilágyi and Balázs Pintér}, 
	title = {Flexcoder: Practical Program Synthesis with Flexible Input Lengths and Expressive Lambda Functions}, 
	booktitle = {arXiv.org}, 
	year = {2021}, 
}
Project image
Flexcoder: Gyakorlati programszintézis flexibilis inputhosszokkal és kifejező lambdafüggvényekkel (Regional: 1st Place; National: 2nd Place)
Bálint Mucsányi, Bálint Gyarmathy, Ádám Czapp
National Conference of Scientific Students' Associations, Hungary (OTDK; in Hungarian), 2021
Paper / Supplemental / Poster /
@InProceedings{Mucsanyi2021OTDK, 
	author = {Bálint Mucsányi and Bálint Gyarmathy and Ádám Czapp}, 
	title = {Flexcoder: Gyakorlati programszintézis flexibilis inputhosszokkal és kifejező lambdafüggvényekkel}, 
	booktitle = {National Conference of Scientific Students' Associations, Hungary (OTDK; in Hungarian)}, 
	year = {2021}, 
}

Miscellaneous

Project image
Analyzing Influential Factors of a Successful Movie
Bálint Mucsányi, Daniel Dauner
Project Work at the University of Tübingen, 2022
Paper / Code /
@InProceedings{Mucsanyi2022TUE, 
	author = {Bálint Mucsányi and Daniel Dauner}, 
	title = {Analyzing Influential Factors of a Successful Movie}, 
	booktitle = {Project Work at the University of Tübingen}, 
	year = {2022}, 
}
Project image
Hatékony példaalapú programszintézis fastruktúrájú nyelvtanon (Best Thesis Award)
Bálint Mucsányi
BSc Thesis at ELTE Eötvös Loránd University (in Hungarian), 2021
Paper / Code /
@InProceedings{Mucsanyi2021BSCTHESIS, 
	author = {Bálint Mucsányi}, 
	title = {Hatékony példaalapú programszintézis fastruktúrájú nyelvtanon}, 
	booktitle = {BSc Thesis at ELTE Eötvös Loránd University (in Hungarian)}, 
	year = {2021}, 
}

Homepage Template

This website is based on the template of Michael Niemeyer. Check out his Github repository for instructions on how to use it.