Komeili is interested in interpretable machine learning, deep neural networks, and transfer learning. The goal is to create ML models with explainability in mind or to develop methods that can decipher existing black-box ML models. He is also interested in how ML models can learn to perform new t...
Komeili is interested in interpretable machine learning, deep neural networks, and transfer learning. The goal is to create ML models with explainability in mind or to develop methods that can decipher existing black-box ML models. He is also interested in how ML models can learn to perform new tasks with a limited amount of labeled data; a capability that human is very good at.