

30167 Hannover


I received my M.Sc. in Applied Statistics and B.Sc. in Economics from the University of Göttingen. In my studies I focused on Machine & Deep Learning, (Bayesian) Generalized Linear Regression methods and Econometrics respectively. My thesis concerned itself with extracting main effects from Bayesian Neural Networks using grouped shrinkage priors and splines; inferring its parameters using Stochastic Gradient Markov Chain Monte Carlo methods.
Since Sep. 2021, I am pursuing my PhD as a member of Prof. Lindauer’s group. My current research interests are Bayesian- & multi-fidelity optimization and meta-learning, aiming at boosting the performance of machine learning algorithms by choosing appropriate hyperparameters in a data driven, principled and efficient manner. Most recently we investigated, how to combine multi-fidelity and meta-learning for algorithm selection using a transformer architecture. Currently I am involved in finding a way of training reinforcement learning agents more robustly as well as training graph-based models more efficiently using a novel fidelity type.
Research Interests
- Bayesian Optimization
- Multi-Fidelity for Hyperparameter Optimization
- Multi-Fidelity for Graph Neural Networks
- Meta-Learning for Hyperparameter Optimization
- Reinforcement Learning for Algorithm Selection on partial learning curves
- Hyperparameter Optimization for Reinforcement learning
Curriculum Vitae
-
Work Experience
2021-today Doctoral Researcher Leibniz University Hannover
2021 (6 Monate) Student Assistant, Georg-August University Göttingen
2019 (6 Monate) Internship: Data Architecture and Smart Analytics, Deutsche Bank AG
2017-2018 (18 Monate) Student Assistant, Georg-August University Göttingen
-
Education
Since 2021 Ph.D. Student in AutoML, Leibniz University Hannover
2017-2020 M.Sc Applied Statistics, Georg-August University Göttingen
2014-2017 B.Sc. Economics, Georg-August University Göttingen
Publications
2022
Mohan, A., Ruhkopf, T., & Lindauer, M. (2022). Towards Meta-learned Algorithm Selection using Implicit Fidelity Information. in ICML Workshop on Adaptive Experimental Design and Active Learning in the Real World (ReALML)
Ruhkopf, T., Mohan, A., Deng, D., Tornede, A., Hutter, F., & Lindauer, M. (2022). MASIF: Meta-learned Algorithm Selection using Implicit Fidelity Information.
Projekte
-
CoyPu: Cognitive Economy Intelligence Platform for the Resillience of Economic EcosystemsNatural disasters, pandemics, financial- & political crises, supply shortages or demand shocks propagate through hidden and intermediate linkages across the global economic system. This is a consequence of the continuous international division of business and labor which is at the heart of globalisation. The aim of the project is to provide a platform that expounds the complex supply chains and reveal the linkages, compounded risks and provide companies with predictions regarding their exposure in various granularities.Leitung: Prof. Marius Lindauer and Prof. Maria Esther-Vidal (L3S/LUH)Team:Jahr: 2021Förderung: Innovationswettbewerb Künstliche Intelligenz (BMWK)Laufzeit: 2021-2024