Unverified50% confidenceFactExact time
Sample efficiency refers to the number of interaction samples an agent needs to reach a certain performance level, and it is one of the core metrics for putting online reinforcement learning into practice.
1
Sources
50%
Confidence
Long-term
Relevance
11/2/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMulti-turn reinforcement learning uses the agent's multi-step interactions in a real environment as training signals, awarding rewards based on the quality of the final retrieval results.76% similarUnverifiedThe MASCAL method lets each agent jointly optimize task performance and confidence calibration during reinforcement learning, learning to generate calibrated correct probabilities for its answers.76% similarUnverifiedReinforcement learning requires massive numbers of interaction samples, so environment running speed and parallelization capability are critical for shortening training cycles73% similarUnverifiedTurning agent skill optimization into an iterative, evaluable engineering workflow helps improve reproducibility and measurability, at lower cost, with faster iteration, and without relying on retraining the base model.73% similarUnverifiedThe effectiveness of imitation learning depends heavily on the quality and quantity of demonstration data, and data quality directly affects the generalization ability of the policy.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1127160API
curl https://kongchang.com/api/v1/knowledge/claims/1127160MCP
get_claim(id=1127160)