Rousslan Fernand Julien Dossa
Scholar

Rousslan Fernand Julien Dossa

Google Scholar ID: ZDWS2CMAAAAJ
Kobe University
Deep LearningReinforcement LearningUnsupervised LearningNeuroscienceRobotics
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
833
 
H-index
7
 
i10-index
6
 
Publications
20
 
Co-authors
0
 
Contact
Emaildosssman@hotmail.frCVOpen ↗TwitterOpen ↗GitHubOpen ↗
Publications
1 items
Improving Low-Cost Teleoperation: Augmenting GELLO with Force
2025
Cited
0
Resume (English only)
Academic Achievements
  • [2021-11] Proposed a Hierarchical World Model (HWM) to improve sample efficiency and performance in model-based RL
  • [2021-11] Created CleanRL, an open-source library with high-quality single-file implementations of deep RL algorithms
  • [2021-08] Conducted empirical investigation on early stopping optimizations in Proximal Policy Optimization (PPO)
  • [2020-09] Developed a hybrid approach combining reinforcement and imitation learning for human-like agents
  • Authored technical blog posts on re-implementations and experiments with SAC, MADE, DDPG, and exploration schemes in DQN
Co-authors
0 total
Co-authors: 0 (list not available)