Scholar
Nikola Milosevic
Google Scholar ID: vGYpn-sAAAAJ
Max Planck Institute for Human Cognitive and Brain Sciences
Citations & Impact
Citations
5
H-index
1
i10-index
0
Publications
8
Co-authors
6
list available
Publications
Resume (English only)
Academic Achievements
- Publications: 'Embedding Safety into RL: A New Take on Trust Region Methods', International Conference on Machine Learning (ICML), 2025; Awards: Master's Thesis Award, University of Applied Sciences Leipzig, 2021.
Research Experience
- Currently conducting doctoral research at the Max Planck Institute for Human Cognitive and Brain Sciences, focusing on the safety and reliability of reinforcement learning.
Education
- Ph.D. Candidate, Max Planck Institute for Human Cognitive and Brain Sciences, Neural Data Science Lab, Sep. 2022 - present; M.S. in Electrical Engineering, University of Applied Sciences Leipzig, Sep. 2019 - Jul. 2021.
Background
- Research Interests: Reliability and real-world application of Reinforcement Learning (RL). Specialization: Safety-critical decision-making, learning from imperfect or limited data (offline RL), and developing new methods for exploration and generalization. Overview: Focused on making reinforcement learning systems more safe, stable, and aligned with real-world goals.
Miscellany
- Contact: milose.nik(at)gmail.com; Personal Website: https://skylerhallinan.com/; Google Scholar and GitHub profiles available.
