Daniel Csillag
Scholar

Daniel Csillag

Google Scholar ID: 8QEfviAAAAAJ
FGV EMAp
Machine LearningStatistics
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
25
 
H-index
2
 
i10-index
1
 
Publications
12
 
Co-authors
0
 
Contact
Emaildaniel.csillag@fgv.brTwitterOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
7 items
Avoid What You Know: Divergent Trajectory Balance for GFlowNets
2026
Cited
0
Random Gradient-Free Optimization in Infinite Dimensional Spaces
2025
Cited
0
Adaptive Training of INRs via Pruning and Densification
2025
Cited
0
Differentially Private E-Values
2025
Cited
0
Extending Prediction-Powered Inference through Conformal Prediction
2025
Cited
0
Image Super-Resolution with Guarantees via Conformal Generative Models
2025
Cited
0
Prediction-Powered E-Values
2025
Cited
0
Resume (English only)
Academic Achievements
  • Paper 'Differentially Private E-Values' published on ArXiv
  • Paper 'Conformal Prediction-Powered Inference' published on ArXiv
  • Paper 'Conformal Super-Resolution' accepted at NeurIPS 2025
  • Developed software tools: Rum (background process management), Tap (abstraction over C/C++ build systems), Logrun (minimalist Python experiment logging)
  • Recent talks include: 'What do we need for models that can reason causally?' at IMPA on February 12, 2025; 'Confidence sequences for general M-estimation' at FGV EMAp on April 02, 2024
Co-authors
0 total
Co-authors: 0 (list not available)