Which type of machine learning involves feedback based on actions?

Reinforcement learning is a type of machine learning where an agent learns to make decisions by taking actions in an environment to maximize cumulative rewards. This type of learning is characterized by the feedback loop it creates: the agent receives positive or negative feedback based on its actions, which influences its future decisions. The goal is to learn a policy that dictates the best action to take in given situations to achieve the highest possible reward over time.

In contrast, supervised learning involves learning from labeled data where the correct output is already known, while unsupervised learning deals with unlabeled data and finds patterns or groupings within the data without predefined outcomes. Collaborative learning, although related to working together to solve problems, does not involve the same action- and feedback-driven mechanism that reinforces learning through rewards. Therefore, reinforcement learning stands out as the correct choice due to its unique focus on learning from the consequences of actions taken in an interactive environment.