Daphne Cornelisse
Scholar

Daphne Cornelisse

Google Scholar ID: ScR5fBYAAAAJ
Graduate student, NYU
multi-agent systemsreinforcement learningimitation learning
Citations & Impact
All-time
Citations
96
 
H-index
4
 
i10-index
4
 
Publications
9
 
Co-authors
1
list available
Resume (English only)
Academic Achievements
  • Selected papers: 'Building reliable sim driving agents by scaling self-play' (under submission), 'GPUDrive: Data-driven, multi-agent driving simulation at 1 million FPS' (ICLR 2025), 'Human-compatible driving partners through data-regularized self-play reinforcement learning' (RLC 2024). Awards: 2025 PhD Cooperative AI Fellowship, Chishiki AI Fellow (funded by NSF).
Research Experience
  • Summer 2025 internship with Waymo's safety research team; gave a talk at Waymo on guided self-play; delivered an invited talk at the AI and Social Good Lab (AISOC) at CMU; co-organized the workshop on Computational Models of Human Road User Behavior for Autonomous Vehicles at IEEE IAVVC.
Education
  • Educational background includes an undergraduate degree in Neuroscience and a master’s in AI, where she worked on cooperative game theory. Currently, she is a PhD student at the EMERGE lab of NYU, advised by Prof. Eugene Vinitsky.
Background
  • Research interests: multi-agent reinforcement learning, imitation learning, and principles from cognitive science. Briefly: She is a third-year PhD student at NYU, focusing on building agents that behave like humans in multi-agent, safety-critical settings.
Miscellany
  • Personal interests not explicitly mentioned.