Dipl-Ing, Dr. rer. nat. Nico Görnitz
- Ex-Postdoc | AI Research | Berlin Center for Machine Learning | TU Berlin, MPI Tübingen, Microsoft Research LA
- Freelancer | AI Consulting and MVP, Coachings, and Workshops | Berlin
- AI Entrepreneur | 123ai.de | Berlin
Anomaly detection, or outlier analysis, has been the main focus of my research career and I published numerous papers, applications and theoretical, on various related topics over the course of 10+ years. I am still frequently serving as a reviewer and program committee member for international journals and conferences.
A short bio
After an internship with the eScience Group, led by David Heckerman (Microsoft Research, Los Angeles, US) in 2014, I received a scholarship and was enrolled as a research associate in the machine learning group at the Berlin Institute of Technology (TU Berlin, Berlin, Germany) headed by Klaus-Robert Müller as part of the center for machine learning (BZML: Berliner Zentrum für Maschinelles Lernen) which I later also joined as a postdoc. I also started my research career in the machine learning group of the TU Berlin in 2008 and was, between 2010 and 2012, also affiliated with the Friedrich Miescher Laboratory of the Max Planck Society in Tübingen, where I was co-advised by Gunnar Rätsch.
I received a diploma degree (MSc equivalent) in computer engineering (Technische Informatik) from the Berlin Institute of Technology with a thesis in machine learning for computer security as well as a PhD (summa cum laude) in machine learning with a thesis entitled “One-class classification in the presence of point, collective, and contextual anomalies”.
Since September 2019, I am a freelancer and AI entrepreneur based in Berlin, Germany.
My research interests
I am interested in all kinds of machine learning and have mainly been working on one-class classification based anomaly detection (aka one-class learning, concept learning, rare class detection, density level-set estimation), large-margin structured output learning, corresponding optimization techniques and ubiquitous deep learning techniques. Applications that I have been working on cover computational biology (DFG funded project on ‘Learning with Sequence Data’), computer security (BMBF funded project ‘ReMIND’), computational sustainability (BMBF funded project ‘ALICE: autonomous learning in complex environments’ I + II) but also brain-computer-interfaces, natural language processing and computational geosciences.
My research activities
I (co-)authored a number of scientific research workshop, conference, and journal articles. For a full list of my publications please refer to my Google Scholar site.
As an organizer, I was responsible for booking locations, selecting and scheduling speakers, and defining the scope of the event:
- ICML Workshop on Anomaly Detection 2016
- IDA Joint Seminar Castle Wulkow 2013, TU Berlin
- MPI Group Retreat Heiligkreuztal 2011
As a guest editor, I was selecting and reviewing a number of manuscripts for a certain topic (Special issue on Sports Analytics):
- DAMI/DMKD 2016
As a member of the program committee, I was responsible for inviting and scheduling speakers as well as reviewing scientific work:
- ACML 2018, 2019
- ECML 2018, 2019
- KDD 2018, 2019
- IJCAI 2018
- IEA/AIE 2017 Special Track on Anomaly Detection
- AAAI 2016-2018
- ICML 2016 Anomaly Detection Workshop
- IDA Seminar Castle Wulkow 2013
- IJCAI 2011
During my scientific career, I was a frequent reviewer of many internationally well-known journals, conferences, and books:
- TNNLS, ICPR, ICVGIP, WSDM, JMLR, TBME, TPAMI
ICML, DAGM, IJCAI, STCO, NCAA, KDD, GCPR, AAAI, PlosOne, MLJ,
IEA/AIE, CSDA, ECML, NIPS, TAAI, MLSP, KAIS, ACML, Bentham Science, etc. pp.
Invited talks and workshops
- One-class Learning based Anomaly Detection for Data with Dependency Structure (Zalando Research Talks, 12. Jan 2017)
- Eine Einführung in die Künstliche Intelligenz (Vorstandstagung der Dt. Pathologen, 23. June 2018)
- Introduction to Anomaly Detection (Petrobras, Brazil, 26.-29. November 2018)
- Introduction to Machine Learning (Petrobras, Brazil, 24.4.-8.5. 2019)
- I attended the machine learning summer school in Bordeaux (2011)
- The nominated representative for BMBF founded ALICE project at the CeBIT trade show (2014)
- (Best Paper Award) Our paper on the explanation of non-linear learning algorithms won the best paper award at the NIPS workshop on interpretability, 2016
During my time at TU Berlin, I held various lectures for MSc students (math, engineering, computer science mainly), e.g.:
- Kernel Methods lecture as part of the Beginner’s Workshop on Machine Learning (Workshop, Winter Term, 2018)
- Applications of Cognitive Algorithms (Seminar, Summer Term 2014)
- Mathematical Foundations of Machine Learning (Block Seminar, Summer Term 2014)
- Classical Topics in Machine Learning (Seminar, Winter Term 2013/2014)
- Mathematical Foundations of Machine Learning (Block Seminar, Winter Term 2013/2014)
- Hot Topics in Machine Learning (Seminar, Summer Term 2013)
- Mathematical Foundations of Machine Learning (Block Seminar, Summer Term 2013)
In addition to faculty teachings, I organized and held workshops for industry and academia, e.g.
- Workshop on Anomaly Detection (1 week, full day, lectures and exercises), Petrobras, Rio de Janeiro, Brazil, 2018
- Introduction to Machine Learning Workshop (2 week, full day, organization and lectures on Bayesian learning and variational inference)
- 2016 ICML Workshop on Anomaly Detection, International Conference on Machine Learning, New York, 2016
I had the honor to work with very smart and independent students which resulted not only in extraordinary theses but also in various peer-reviewed publications. Many of the students below continued to become successful researchers.
- Towards the Explanation of Deep Learning Methods on CRISPR/Cas Applications (S. Proft, internship, 2018)
- Graph-Based Anomaly Detection: Finding Sybil Networks with Supervised Random Walks (J. Höhner, master thesis, 2017)
- Predicting the Development of Financial Markets by Applying Machine Learning Tools to Social Media Data (Matthias Manhertz, master thesis, 2016)
- Efficient Algorithms for Exact and Approximate Inference in Sequence Labeling SVMs (Alexander Bauer, master thesis, 2012)
- Analysis of Learning Models for Motif Recognition in DNA Sequences (M. M.-C. Vidovic, master thesis, 2012)