A robotics engineering team is training a warehouse picking robot in a simulated environment. The robot receives a positive score each time it retrieves an item and returns it to the correct bin within a time limit, and a negative score whenever it collides with a shelf or drops an item. Over thousands of simulated attempts, the robot's navigation strategy improves without any human labeling individual moves as right or wrong. Which type of machine learning approach is being used?
Select an answer to reveal the explanation.
Short Explanation
Think about training a dog with treats and a firm no rather than handing it a rulebook. Nobody is marking each individual paw movement right or wrong here. The robot just gets a good score when it finishes the job cleanly and a bad score when it crashes into something, and after enough tries it starts favoring the paths that tend to earn the good score. That is the trial-and-reward loop at the heart of reinforcement learning: an agent, an environment, and a running score it is trying to maximize over time. Contrast that with an approach where someone hand-labels thousands of examples as correct or incorrect before training even starts, which needs a labeled answer key up front. It also is not the same as sorting warehouse layouts into groups by how similar they look, since that never involves a reward at all. And it is not about predicting a single distance number either, since nothing here is guessing a continuous value from fixed inputs.
Full Explanation
The correct answer is A. The robot is not being told which specific action is correct or incorrect at each step; instead it experiences positive and negative scores that accumulate over an entire simulated attempt, and it adjusts its strategy over many episodes to maximize that cumulative reward. This trial-and-reward pattern, where an agent interacts with an environment and gradually improves its behavior from feedback signals rather than labeled examples, defines reinforcement learning. Option B is incorrect because supervised learning requires a training dataset where a human has already labeled each example with the correct output; here no one is labeling individual robot movements as right or wrong. Option C is incorrect because unsupervised clustering groups data points by similarity in the absence of any feedback signal, while this robot's behavior is explicitly shaped by reward and penalty scores tied to outcomes. Option D is incorrect because regression is a supervised technique for predicting a continuous numeric value from labeled input-output pairs, not a description of an agent improving a multi-step strategy through accumulated reward over time.