Scaling Laws

How do different scaling laws apply in different contexts? I’m going to spend a lot of time understanding and thinking about scaling laws and how they apply to Reinforcement Learning.

There’s Moore Law.

Resources:

• https://cameronrwolfe.substack.com/p/llm-scaling-laws
• https://medium.com/sage-ai/demystify-transformers-a-comprehensive-guide-to-scaling-laws-attention-mechanism-fine-tuning-fffb62fc2552

Papers:

• Scaling Laws for Neural Language Models

Neural Scaling Law

https://en.wikipedia.org/wiki/Neural_scaling_law

For RL, see Scaling RL.

Robotics Scaling Laws

https://agentic.substack.com/p/whats-missing-for-robot-foundation