Michael Herman
Research scientist


Michael Herman
Date of birth:
May 05, 1987


Currently, I am a Project Lead and Research Scientist working on Behavior Prediction for Assisted and Automated Driving at the Bosch Center for Artificial Intelligence. Together with my team, I am solving Human Motion Prediction Problems using Deep Learning, Probabilistic Graphical Models, and Imitation Learning. Prior to that I did my PhD at Bosch and the Autonomous Intelligent Systems Lab at the University of Freiburg, which is headed by Prof. Wolfram Burgard. My current research is focused on learning prediction models in multi-agent systems. Furthermore, I am interested in learning generalizable representations of an experts motivation or goal from observed behavior to enable programming by demonstration in complex, unknown environments.
I received my Bachelor's degree in Mechatronics and Electrical Engineering at the University of Applied Sciences Esslingen and wrote my Bachelor thesis at Daimler on the computation of Flexray clock precisions of new physical network configurations. For my Master's degree, I studied Electrical Engineering and Information Technology at the University of Stuttgart. My interest in the field of machine learning has been aroused at the Institute of Signal Processing and System Theory, where I wrote a study about Speech Emotion Recognition using Hidden Markov Models and my Master's thesis on Active Learning for Image Retrieval.


Machine Learning
Artificial Intelligence
Deep Learning
Probabilistic Inference
Learning from Demonstration
Inverse Reinforcement Learning
Reinforcement Learning
Cognitive Robotics


2013 - 2020: PhD in Computer Science (University of Freiburg / Robert Bosch GmbH)

Thesis: "Simultaneous Estimation of Rewards and Dynamics in Inverse Reinforcement Learning Problems"

2011 - 2013: M.Sc. Electrical Engineering and Information Technology (University of Stuttgart)

Thesis: "Active Learning for Image Retrieval"

2007 - 2011: B.Eng. Mechatronics and Electrical Engineering (University of Applied Sciences Esslingen)

Thesis: "Development of an AUTOSAR-Based FlexRay Test Setup for Determining a Minimum Configuration for the Dual Channel Mode of FlexRay"


2019/06 - present: Project Lead & Research Scientist at Bosch Center for Artificial Intelligence (Renningen, Germany)

Learning to make predictions of multi-agent human behavior for assisted and automated driving (Imitation Learning, Deep Learning, Probabilistic Graphical Models)

2008/05 - present: Executive partner of Herman & Kegreiß GbR (Böblingen, Germany)

Project management, search engine optimization, software development

2016/10 - 2019/06: Research Scientist at Bosch Center for Artificial Intelligence (Renningen, Germany)

Learning to make predictions in multi-agent environments (Imitation Learning, Deep Learning)

2013/08 - 2016/07: PhD student at Robert Bosch GmbH (Renningen, Germany)

Simultaneous estimation of rewards and dynamics in Inverse Reinforcement Learning problems

2011/03 - 2013/05: Student employee at EVOMOTIV GmbH (Leinfelden-Echterdingen, Germany)

Hardware development of electrical circuit boards, embedded software development, signal processing

2010/08 - 2011/02: Bachelor thesis at Daimler AG (Sindelfingen, Germany)

Computation of Flexray clock precisions of new physical network configurations, development of an AUTOSAR-based test setup for determining a minimum configuration for the dual channel mode of FlexRay

2009/09 - 2010/02: Internship at Daimler AG (Böblingen, Germany)

Development and implementation of an automatic configuration of vehicle data logger systems

2007/07 - 2007/09: Internship at Siemens AG (Weilimdorf, Germany)

Fundamentals of electrical engineering



Python, MATLAB, C#, C++, C, HTML, CSS, JavaScript, PHP, SQL

Other interests:

Guitar, piano, singing, squash, snowboarding, electronics