Visual Question Answering

Trained a Visual Question Answering model that takes an image and a natural language question as input and predicts the answer. Uses attention mechanisms to focus on relevant image regions and transfer learning for improved performance.