Nicolai Dorka

Researcher, Amazon

Joined

September 2019

Names

Nicolai Dorka (Preferred)

Emails

****@informatik.uni-freiburg.de (Confirmed)

,

****@cs.uni-freiburg.de (Confirmed)

,

****@gmail.com (Confirmed)

Personal Links

Career & Education History

Researcher

Amazon (amazon.co.uk)

2025 – Present

PhD student

University of Freiburg, Universität Freiburg (cs.uni-freiburg.de)

2018 – 2024

Advisors, Relations & Conflicts

PhD Advisor

Wolfram Burgard

Present

Expertise

Deep Learning

Present

Reinforcement Learning

Present

Publications

Quantile Regression for Distributional Reward Models in RLHF
Nicolai Dorka
- ICLR 2025 Conference Withdrawn Submission
- Readers: Everyone
What Matters in Employing Vision Language Models for Tokenizing Actions in Robot Control?
Nicolai Dorka, Chenguang Huang, Tim Welschehold, Wolfram Burgard
- VLMNM 2024
- Readers: Everyone
Dynamic Update-to-Data Ratio: Minimizing World Model Overfitting
Nicolai Dorka, Tim Welschehold, Wolfram Burgard
- Published: 01 Feb 2023, Last Modified: 14 Jan 2026
- ICLR 2023 poster
- Readers: Everyone
Dynamic Update-to-Data Ratio: Minimizing World Model Overfitting
Nicolai Dorka, Tim Welschehold, Wolfram Burgard
- 28 May 2022 (modified: 30 Dec 2025)
- DARL 2022
- Readers: Everyone
Adaptively Calibrated Critic Estimates for Deep Reinforcement Learning
Nicolai Dorka, Joschka Boedecker, Wolfram Burgard
- 12 Oct 2021 (modified: 12 Oct 2025)
- Deep RL Workshop NeurIPS 2021
- Readers: Everyone

Co-Authors