Contents

New Insights for Scaling Laws in Autonomous Driving

Contents

AI performance have been powered by scale.

500,000 hours of driving

  • Similar to LLMs, motion forecasting quality also follows a power-law as a function of training compute.
  • Data scaling is critical for improving the model performance.
  • Scaling inference compute also improves the model’s ability to handle more challenging driving scenarios.
  • Closed-loop performance follows a similar scaling trend. This suggests, for the first time, that real-world AV performance can be improved by increasing training data and compute.

References