sample-efficiently
Learning a task from relatively few examples; larger models tend to be more sample-efficient.
As defined in Is RL + LLMs enough for AGI? — Sholto Douglas & Trenton Bricken
Where it comes up · 1
Most important claims · 1 of 1Search all claims for “sample-efficiently” →
- Current models are likely still smaller than the human brain—a ~2 trillion parameter model versus an estimated 30 to 300 trillion synapses—which matters because larger models learn more sample-efficiently with fewer demos and have more capacity to form deep, broad generalizations rather than shallow imitation.Is RL + LLMs enough for AGI? — Sholto Douglas & Trenton Bricken