Dr.-Ing. Jan Melchior

RUB
Computer Science
INI
Alumni
Dr.-Ing. Jan Melchior

jan.melchior@ini.rub.de (+49) 234-32-27987

Theory of Neural Systems

(+49) 234-32-27987

I am interested in unsupervised machine learning especially unsupervised 'probabilistic' models, which involves Boltzmann machines and deep learning. Since unsupervised deep networks are hard to train I had to face the question of how they can be trained more successfully. A solid understanding of how these networks model the probability distribution underlying the data is an important aspect and has been besides improved training methods in focus of my research. Other important topics that were in focus in the second half of my PhD are the idea of developmental-online learning, growing systems, and semi supervised machine learning, which I believe are important topics for successful generalization and knowledge integration.

Work Experience

2012 - present

Doctoral Researcher
Doctoral Researcher in the group of Prof. Dr. Laurenz Wiskott at the Institute for Neural Computation, Ruhr-University Bochum (Germany).
2011

Easter school
Computer Vision and Pattern Recognition Easter school 2011, Kioloa, ACT, Australia
2011 - 2012

Internship (6 month)
Internship in the group of Prof. Dr. Marcus Hutter at the ANU College of Engineering and Computer Science, Australian National University, Canberra. (Australia)
2008 - present

Working student
Part time IT-Assistant at AUTOonline, Neuss (Germany).
2008 - 2012

Working student
Part time IT-Assistant at Control€xpert, Langenfeld Rhld. (Germany)
2005 - 2008

Working Student
Part time job as postman at Deutsche Post AG, Langenfeld Rhld. (Germany)
2004

Internship (3 month)
Internship at Dücker Fördertechnik, Langenfeld Rhld. (Germany)

Education

2009 - 2012

Master of Computer Science
Ruhr University Bochum (Germany)

Thesis: Learning Natural Image Statistics with Gaussian-Binary Restricted Boltzmann Machines.
2005 - 2009

Bachelor of Computer Science.
Thesis: Implementierung von Partikel- Schwarm-Optimierung und Vergleich mit einer Evolutionsstrategie
1993 - 2004

Abitur (German general qualification for university entrance)
Konrad Adenauer Gymnasium, Langenfeld Rhld. (Germany)

Teaching

SS18

Thesis supervisor
Biermann, J., 11.10.2018. Predicting thermal lifts from gliding data using artificial neural networks. Bachelor’s thesis, Applied Informatics, Univ. of Bochum, Germany.
WS17

Teaching assistant
Machine Learning: Unsupervised Methods
SS17

Lab course
Scientific Computing with Python
SS16

Lab course
Scientific Computing with Python
SS15

Lab course
Scientific Computing with Python
SS15

Thesis supervisior
Tiesetskaya, Y., 09.06.2015. Implementierung und Vergleich von hierarchischen Auto Encoder Netzwerken. Bachelor’s thesis, Applied Informatics, Univ. of Bochum, Germany.
WS15

Thesis supervisior
Marquardt, Y, 03.03.2015. Implementing Hierachical Sparse Coding and Comparing it to other Hierachical Models. Matser’s thesis, Applied Informatics, Univ. of Bochum, Germany.
WS14

Thesis supervisior
Bölükbasi, K, 23.01.2014. Implementation and Evaluation of Autoencoders. Matser’s thesis, Applied Informatics, Univ. of Bochum, Germany.
SS13

Study project supervisior
Bölukbasi, K., Kastrau, S. 20.06.2013. Implementing Feed Forward Neural Networks with Python. Study project, Applied Informatics, Univ. of Bochum, Germany.

2024

Hebbian Descent: A Unified View on Log-Likelihood Learning

Melchior, J., Schiewer, R., & Wiskott, L.

Neural Computation, 36(9), 1669–1712

@article{MelchiorSchiewerWiskott2024,
	author		=	{Melchior, Jan and Schiewer, Robin and Wiskott, Laurenz},
	title		=	{Hebbian Descent: A Unified View on Log-Likelihood Learning},
	journal		=	{Neural Computation},
	volume		=	{36},
	number		=	{9},
	pages		=	{1669–1712},
	month		=	{August},
	year		=	{2024},
	doi		=	{10.1162/neco_a_01684},
}

A neural network model for online one-shot storage of pattern sequences
Melchior, J., Altamimi, A., Bayati, M., Cheng, S., & Wiskott, L.
PLOS ONE, 19(6), 1–28

link

@article{MelchiorAltamimiBayatiEtAl2024, author = {Melchior, Jan and Altamimi, Aya and Bayati, Mehdi and Cheng, Sen and Wiskott, Laurenz}, title = {A neural network model for online one-shot storage of pattern sequences}, journal = {PLOS ONE}, volume = {19}, number = {6}, pages = {1–28}, month = {June}, year = {2024}, doi = {10.1371/journal.pone.0304076}, }

Melchior, J., Altamimi, A., Bayati, M., Cheng, S., & Wiskott, L.. (2024). A neural network model for online one-shot storage of pattern sequences. PLOS ONE, 19(6), 1–28. http://doi.org/10.1371/journal.pone.0304076

2021

Fully Automated, Realistic License Plate Substitution in Real-Life Images
Kacmaz, U., Melchior, J., Horn, D., Witte, A., Schoenen, S., & Houben, S.
In Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC) (pp. 2972–2979)

Paper

@inproceedings{KacmazMelchiorHornEtAl2021, author = {Kacmaz, Ufuk and Melchior, Jan and Horn, Daniela and Witte, Andreas and Schoenen, Sebastian and Houben, Sebastian}, title = {Fully Automated, Realistic License Plate Substitution in Real-Life Images}, booktitle = {Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC)}, pages = {2972–2979}, year = {2021}, }

Kacmaz, U., Melchior, J., Horn, D., Witte, A., Schoenen, S., & Houben, S.. (2021). Fully Automated, Realistic License Plate Substitution in Real-Life Images. In Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC) (pp. 2972–2979).

2019

A Hippocampus Model for Online One-Shot Storage of Pattern Sequences
Melchior, J., Bayati, M., Azizi, A., Cheng, S., & Wiskott, L.
CoRR e-print arXiv:1905.12937

@misc{MelchiorBayatiAziziEtAl2019, author = {Melchior, Jan and Bayati, Mehdi and Azizi, Amir and Cheng, Sen and Wiskott, Laurenz}, title = {A Hippocampus Model for Online One-Shot Storage of Pattern Sequences}, howpublished = {e-print arXiv:1905.12937}, year = {2019}, }

Melchior, J., Bayati, M., Azizi, A., Cheng, S., & Wiskott, L.. (2019). A Hippocampus Model for Online One-Shot Storage of Pattern Sequences. CoRR. e-print arXiv:1905.12937. Retrieved from https://arxiv.org/abs/1905.12937

2018

Storage fidelity for sequence memory in the hippocampal circuit

Bayati, M., Neher, T., Melchior, J., Diba, K., Wiskott, L., & Cheng, S.

PLOS ONE, 13(10), e0204685

@article{BayatiNeherMelchiorEtAl2018,
	author		=	{Bayati, Mehdi and Neher, Torsten and Melchior, Jan and Diba, Kamran and Wiskott, Laurenz and Cheng, Sen},
	title		=	{Storage fidelity for sequence memory in the hippocampal circuit},
	journal		=	{PLOS ONE},
	volume		=	{13},
	number		=	{10},
	pages		=	{e0204685},
	month		=	{October},
	year		=	{2018},
	doi		=	{10.1371/journal.pone.0204685},
}

Utilizing Slow Feature Analysis for Lipreading
Freiwald, J., Karbasi, M., Zeiler, S., Melchior, J., Kompella, V., Wiskott, L., & Kolossa, D.
In Speech Communication; 13th ITG-Symposium (pp. 191–195) VDE Verlag GmbH

@inproceedings{FreiwaldKarbasiZeilerEtAl2018, author = {Freiwald, J. and Karbasi, M. and Zeiler, S. and Melchior, J. and Kompella, V. and Wiskott, L. and Kolossa, D.}, title = {Utilizing Slow Feature Analysis for Lipreading}, booktitle = {Speech Communication; 13th ITG-Symposium}, pages = {191–195}, publisher = {VDE Verlag GmbH}, month = {October}, year = {2018}, }

Freiwald, J., Karbasi, M., Zeiler, S., Melchior, J., Kompella, V., Wiskott, L., & Kolossa, D. (2018). Utilizing Slow Feature Analysis for Lipreading. In Speech Communication; 13th ITG-Symposium (pp. 191–195). VDE Verlag GmbH. Retrieved from https://ieeexplore.ieee.org/document/8578021

2017

Gaussian-binary restricted Boltzmann machines for modeling natural image statistics

Melchior, J., Wang, N., & Wiskott, L.

PLOS ONE, 12(2), 1–24

@article{MelchiorWangWiskott2017,
	author		=	{Melchior, Jan and Wang, Nan and Wiskott, Laurenz},
	title		=	{Gaussian-binary restricted Boltzmann machines for modeling natural image statistics},
	journal		=	{PLOS ONE},
	volume		=	{12},
	number		=	{2},
	pages		=	{1–24},
	month		=	{February},
	year		=	{2017},
	doi		=	{10.1371/journal.pone.0171015},
}

Generating sequences in recurrent neural networks for storing and retrieving episodic memories
Bayati, M., Melchior, J., Wiskott, L., & Cheng, S.
In Proc. 26th Annual Computational Neuroscience Meeting (CNS*2017): Part 2

@inproceedings{BayatiMelchiorWiskottEtAl2017, author = {Bayati, Mehdi and Melchior, Jan and Wiskott, Laurenz and Cheng, Sen}, title = {Generating sequences in recurrent neural networks for storing and retrieving episodic memories}, booktitle = {Proc. 26th Annual Computational Neuroscience Meeting (CNS*2017): Part 2}, year = {2017}, }

Bayati, M., Melchior, J., Wiskott, L., & Cheng, S.. (2017). Generating sequences in recurrent neural networks for storing and retrieving episodic memories. In Proc. 26th Annual Computational Neuroscience Meeting (CNS*2017): Part 2. Retrieved from http://europepmc.org/articles/PMC5592442

2016

How to Center Deep Boltzmann Machines
Melchior, J., Fischer, A., & Wiskott, L.
Journal of Machine Learning Research, 17(99), 1–61

Journal article page

@article{MelchiorFischerWiskott2016, author = {Melchior, Jan and Fischer, Asja and Wiskott, Laurenz}, title = {How to Center Deep Boltzmann Machines}, journal = {Journal of Machine Learning Research}, volume = {17}, number = {99}, pages = {1–61}, year = {2016}, }

Melchior, J., Fischer, A., & Wiskott, L.. (2016). How to Center Deep Boltzmann Machines. Journal of Machine Learning Research, 17(99), 1–61. Retrieved from http://jmlr.org/papers/v17/14-237.html

2014

Gaussian-binary Restricted Boltzmann Machines on Modeling Natural Image statistics
Wang, N., Melchior, J., & Wiskott, L.
(Vol. 1401.5900) arXiv.org e-Print archive

Gaussian-binary Restricted Boltzmann Machines on Modeling Natural Image statistics

@techreport{WangMelchiorWiskott2014, author = {Wang, Nan and Melchior, Jan and Wiskott, Laurenz}, title = {Gaussian-binary Restricted Boltzmann Machines on Modeling Natural Image statistics}, year = {2014}, }

Wang, N., Melchior, J., & Wiskott, L.. (2014). Gaussian-binary Restricted Boltzmann Machines on Modeling Natural Image statistics (Vol. 1401.5900). arXiv.org e-Print archive. Retrieved from http://arxiv.org/abs/1401.5900

2013

How to Center Binary Restricted Boltzmann Machines
Melchior, J., Fischer, A., Wang, N., & Wiskott, L.
(Vol. 1311.1354) arXiv.org e-Print archive

How to Center Binary Restricted Boltzmann Machines

@techreport{MelchiorFischerWangEtAl2013, author = {Melchior, Jan and Fischer, Asja and Wang, Nan and Wiskott, Laurenz}, title = {How to Center Binary Restricted Boltzmann Machines}, year = {2013}, }

Melchior, J., Fischer, A., Wang, N., & Wiskott, L.. (2013). How to Center Binary Restricted Boltzmann Machines (Vol. 1311.1354). arXiv.org e-Print archive. Retrieved from http://arxiv.org/pdf/1311.1354.pdf

2012

An Analysis of Gaussian-Binary Restricted Boltzmann Machines for Natural Images
Wang, N., Melchior, J., & Wiskott, L.
In Proc. 20th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Apr 25–27, Bruges, Belgium (pp. 287–292)

An Analysis of Gaussian-Binary Restricted Boltzmann Machines for Natural Images

@inproceedings{WangMelchiorWiskott2012, author = {Wang, Nan and Melchior, Jan and Wiskott, Laurenz}, title = {An Analysis of Gaussian-Binary Restricted Boltzmann Machines for Natural Images}, booktitle = {Proc. 20th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Apr 25–27, Bruges, Belgium}, pages = {287–292}, year = {2012}, }

Wang, N., Melchior, J., & Wiskott, L.. (2012). An Analysis of Gaussian-Binary Restricted Boltzmann Machines for Natural Images. In Proc. 20th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Apr 25–27, Bruges, Belgium (pp. 287–292).

PyDeep is a machine learning library I created during my PhD. It is written purely in Python/numpy and has a focus on unsupervised machine learning such as PCA, ICA, RBM, AE, ... It contains the code I used in my publications allowing you to reproduce the experiments.

→ PyDeep on GitHub

→ PyDeep Documentation

Winter Term 2017/2018

Lectures

Machine Learning: Unsupervised Methods

Summer Term 2017

Lab courses

Scientific Computing with Python

Summer Term 2016

Lab courses

Scientific Computing with Python

Summer Term 2015

Lab courses

Scientific Computing with Python

About the INI

The Institut für Neuroinformatik (INI) is a research unit of the Faculty of Computer Science at the Ruhr-Universität Bochum. Its scientific goal is to understand the fundamental principles through which organisms generate behavior and cognition while linked to their environments through sensory and effector systems. Inspired by our insights into such natural cognitive systems, we seek new solutions to problems of information processing in artificial cognitive systems. We draw from a variety of disciplines that include experimental psychology and neurophysiology as well as machine learning, neural artificial intelligence, computer vision, and robotics.

Contact

Universitätsstr. 150, Building NB, Room 3/32
D-44801 Bochum, Germany

Tel: (+49) 234 32-28967
Fax: (+49) 234 32-14210

sekretariat@ini.rub.de
www.ini.rub.de