Learning Agent
Domain: Agent Architectures
🎯 Motivation
An agent that improves over time by learning from its experiences - successes, failures, and feedback.
📋 Concrete Example
First time: agent writes code with bugs. User says ‘this has a bug.’ Agent learns: ‘check for null pointers.’ Next time: code is cleaner. Over many interactions, it gets better.
🔗 Analogy
A new employee who learns from feedback - their first report has mistakes. A colleague points them out. Next report is better. Over months, they improve. Learning agents do this automatically.