Reward Signal

إشارة المكافأة

القيمة العددية التي يتلقاها الوكيل بعد كل فعل في التعلم المعزز، تُوجِّهه نحو السلوك المرغوب وتبعده عن غير المرغوب.

The numerical value an agent receives after each action in reinforcement learning, guiding it toward desired behavior and away from undesired behavior.

Also translated asإشارة التعزيز، دالة المكافأة

First appears in this corpus in: Between MDPs and Semi-MDPs — A Framework for Temporal Abstraction in Reinforcement Learning (1999)

Appears in these papers

Constitutional AI: Harmlessness from AI Feedback2022 in the sky ✦
Learning to Summarize from Human Feedback2020 in the sky ✦
Learning to Reason with LLMs2024 in the sky ✦
Between MDPs and Semi-MDPs — A Framework for Temporal Abstraction in Reinforcement Learning1999 in the sky ✦