← Jargon

Jargontechnology· 5 claims· 3 episodes

LoRa

LoRa fine-tuning, a compute-efficient technique for adapting a pre-trained model to new data without full retraining.

As defined in 280 | François Chollet on Deep Learning and the Meaning of Intelligence


Other definitions
  • Low-Rank Adaptation: a fine-tuning method that updates only two small added weight matrices (a delta) instead of the full model, making customization cheaper but learning and forgetting less.

    Ai Sota #2026

Where it comes up · 3

Most important claims · 5 of 5Search all claims for “LoRa” →