Glossary · Term

sample efficiency

← all terms

Definition

Plain language

How much a system can learn from a small amount of data or few attempts.

As stated in the literature

The degree to which a training method reaches good performance with few examples or rollouts; targeted per-step rewards are reported to improve it severalfold over end-of-task-only signals.

Also called: sample-efficient, sample-efficiency

Why it matters: Higher sample efficiency means less data, time, and compute to reach good performance, which makes training cheaper and faster.

For example, one training method might master a task after a handful of attempts while another needs thousands.

Heard on the show

“The nice side effect: it made training something like five to six times more sample-efficient.”
Episode 104 — How Making a Research Agent Smarter Quietly Makes It Leak Your Secrets

Mentioned in 4 episodes

  1. 104
    How Making a Research Agent Smarter Quietly Makes It Leak Your Secrets
  2. 065
    One Loop to Optimize Them All: A Universal API for LLM-Driven Discovery
  3. 053
    An AI Agent Swapped In Focal Loss And Beat A Human-Tuned Training Script
  4. 022
    Training the Model Spec Directly: An Alignment Lever Aimed at the Say-Do Gap

Related terms