Preston Fu
Scholar

Preston Fu

Google Scholar ID: VjtRFvEAAAAJ
UC Berkeley
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
11
 
H-index
2
 
i10-index
0
 
Publications
3
 
Co-authors
0
 
Contact
TwitterOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
3 items
Long-Horizon Language Model Reinforcement Learning via Progressive Point Matching
2026
Cited
0
Compute-Optimal Scaling for Value-Based Deep RL
2025
Cited
0
Value-Based Deep RL Scales Predictably
2025
Cited
0
Resume
Academic Achievements
  • Published several papers including 'Compute-Optimal Scaling for Value-Based Deep RL' (NeurIPS 2025) and 'Value-Based Deep RL Scales Predictably' (ICML 2025), with presentations at NeurIPS Math-AI Workshop (2024) and EMNLP Future Event Detection Workshop (2024).
Research Experience
  • Part of the Accel Scholar program; involved in multiple research projects on reinforcement learning.
Education
  • Attending UC Berkeley, advised by Sergey Levine and Aviral Kumar.
Background
  • An undergrad at UC Berkeley, interested in building systems that can efficiently learn general skills.
Miscellany
  • GitHub: @prestonfu, Linkedin: @preston-fu, Scholar: click, Email: prestonfu [at] berkeley [dot] edu