Dan Haramati
Scholar

Dan Haramati

Google Scholar ID: NrfsdRYAAAAJ
Brown University
Reinforcement LearningGenerative ModelsObject-centric LearningRobotics
Citations & Impact
All-time
Citations
23
 
H-index
2
 
i10-index
1
 
Publications
3
 
Co-authors
4
list available
Resume (English only)
Academic Achievements
  • Publications: - EC-Diffuser: Multi-object Manipulation via Entity-centric Behavior Generation, ICLR 2025; - Entity-centric Reinforcement Learning for Object Manipulation from Pixels, ICLR 2024 Spotlight (top 5%); - Workshop Paper: Decision-making with Deep Latent Particles accepted to the GenAI4DM Workshop at ICLR 2024; - ECRL accepted to the GCRL Workshop at NeurIPS 2023 as a Spotlight Talk.
Research Experience
  • Current Research Focus: World modeling and model-based RL, exploring questions such as how to learn transferable world models, what level of abstraction is suitable for world modeling, and when learning a world model is fundamentally preferable to model-free approaches.
Education
  • PhD: Department of Computer Science at Brown University, Advisor: Prof. George Konidaris; M.Sc.: Electrical and Computer Engineering at the Technion - Israel Institute of Technology, Advisor: Prof. Aviv Tamar; B.Sc.: Electrical Engineering and Physics at the Technion - Israel Institute of Technology.
Background
  • Research Interests: Generalization and transfer in Reinforcement Learning, particularly compositional generalization; Field: Computer Science; Summary: Aiming to build general and adaptive sequential decision-making agents.
Miscellany
  • Interests: Hiking, snowboarding, basketball, and traveling; Prior to academic journey: Spent two years traveling in East Asia, Europe, and America.