Also known as: rl · reinforcement learning (RL) · RL
L’apprentissage par renforcement (RL) est un paradigme d’apprentissage automatique dans lequel un agent apprend à agir dans un environnement en recevant des récompenses ou des pénalités pour ses actions. L’objectif est d’apprendre une politique qui maximise la récompense cumulée attendue au fil du temps. Le RL est utilisé pour des problèmes de décision séquentielle tels que la robotique, les jeux, les systèmes de contrôle et certaines techniques d’alignement de l’IA.
An area of machine learning concerned with how software agents ought to take actions in an environment so as to maximize some notion of cumulative reward. Reinforcement learning is one of three basic machine learning paradigms, alongside supervised and unsupervised learning. It differs from supervised learning in that labelled input/output pairs need not be presented, and sub-optimal actions need not be explicitly corrected. Instead the focus is finding a balance between exploration (of uncharted territory) and exploitation (of current knowledge).
A way of training machine learning systems for a specific application. An AI system is trained by being rewarded for following certain 'correct' strategies and punished if it follows the 'wrong' strategies. After completing a task, the AI system receives feedback, which can sometimes be given by humans (known as 'reinforcement learning from human feedback'). In the feedback, positive values are assigned to 'correct' strategies to encourage the AI system to use them, and negative values are assigned to 'wrong' strategies to discourage them, with the classification of 'correct' and 'wrong' depending on a pre-established outcome. This type of learning is useful for tweaking an AI model to follow certain 'correct' behaviours, such as fine-tuning a chatbot to output a preferred style, tone or format of language (PB 57). See also supervised learning, unsupervised learning and semi-supervised learning.