Ethan Caballero
Papers
1
Total Citations
15
H-Index
1
About
Ethan Caballero is a researcher at the forefront of understanding the fundamental scaling behaviors of deep neural networks. His primary research areas center on neural scaling laws, AI safety, and the theoretical underpinnings of large-scale machine learning systems. Caballero’s most notable contribution is his work on "Broken Neural Scaling Laws" (2022), which introduced a smoothly broken power law functional form that accurately models and extrapolates how evaluation metrics vary with compute during training or inference. This breakthrough provides a more nuanced and precise framework than traditional scaling laws, enabling researchers to predict model performance across different regimes. With 15 citations, this work has already influenced how the AI community thinks about scaling, particularly in identifying where gains from additional compute diminish. Caballero’s research is critical for designing more efficient and predictable neural networks, and his insights are shaping the next generation of large-scale AI systems. He continues to push the boundaries of our understanding of deep learning’s empirical properties, making him a key voice in the ongoing dialogue about AI development and safety.
Research Focus
Key Achievements
Top Papers
- 1Broken Neural Scaling Laws15 citations · 2022