|
Research
I am a member of Alexandra Carpentier's research group concerned with the project “Multi- System Anomaly Detection”.
The third member of the group, Andrea Locatelli, is also a PhD student. For details, visit Alexandra Carpentier's homepage.
My main field of interest is Minimax Composite-Composite Hypothesis Testing.
Teaching
- Since 2010: Many exercise sessions in essentially all basic courses.
- 2014-2017: Elementary courses (lectures+exercises) on Analysis, Linear Algebra and Probability Theory for future teachers.
Talks
-
On “Minimax Euclidean Separation Rates for Testing Convex Hypotheses in Rd”:
- 21 June 2016 at WIAS Berlin, see here.
- 6 July 2016 at Uni Potsdam, PhD Day of the Institute.
- 28 Sept 2016 at “Young Researchers' Meeting in Mathematical Statistics”, HU Berlin, see here.
-
On “Testing the Equality in Distribution of Two Random Graphs”:
- 31 Jan 2017 at WIAS Berlin, see here.
Attended Events
Publications
- A. Locatelli, M. Gutzeit, A. Carpentier. “An optimal algorithm for the Thresholding Bandit Problem.”
Proceedings of The 33rd International Conference on Machine Learning, pp. 1690–1698, 2016. arXiv.
- A. Carpentier, G. Blanchard, M. Gutzeit. “Minimax Euclidean Separation Rates for Testing Convex Hypotheses in Rd.”
(submitted) arXiv preprint 1702.03760.
- D. Ghoshdastidar, M. Gutzeit, A. Carpentier, U. von Luxburg. “Two-Sample Tests for Large Random Graphs
Using Network Statistics.”
COLT 2017. arXiv.
- D. Ghoshdastidar, M. Gutzeit, A. Carpentier, U. von Luxburg. “Two-sample Hypothesis Testing for Inhomogeneous Random Graphs.”
(submitted) arXiv preprint 1707.00833.
|
|