1. Supervised Learning: In this approach, an AI system is trained on a dataset that is labeled with the correct output. The system learns to make predictions by generalizing from the labeled examples. For example, a supervised learning algorithm can be trained to recognize images of cats by being shown a large number of images labeled as "cat" and learning to associate certain features with the label.

2. Unsupervised Learning: In unsupervised learning, the AI system is given a dataset without any labels. The goal is for the system to discover patterns and relationships in the data on its own. This can be useful for tasks such as clustering or dimensionality reduction. For example, an unsupervised learning algorithm can be used to group similar products together based on their features.

3. Reinforcement Learning: In reinforcement learning, an AI system learns by interacting with an environment and receiving feedback in the form of rewards or punishments. The system learns to take actions that maximize its cumulative reward over time. This approach is often used in tasks such as playing games or controlling robots.

4. Transfer Learning: Transfer learning is a technique where an AI system leverages knowledge learned from one task to perform another, similar task. For example, a neural network that has been trained on a large dataset of images can be fine-tuned on a smaller dataset for a specific image classification task.

These are just a few examples of how AI systems can learn. Different learning algorithms and techniques can be combined or modified to suit the specific requirements of a particular task.