MLMD is an AI platform designed to accelerate the discovery of advanced materials by providing a programming-free, user-friendly interface for materials design. The platform integrates various functionalities, including data analysis, descriptor refactoring, hyper-parameter optimization, and property prediction. It supports end-to-end materials design, from data collection and preprocessing to model inference and surrogate optimization, even in scenarios with limited data through active learning. MLMD's key features include: 1. **User-Friendly Interface**: No programming required, making it accessible to researchers without extensive technical backgrounds. 2. **End-to-End Materials Design**: Capable of discovering novel materials with desired properties through a seamless workflow. 3. **Surrogate Optimization**: Utilizes well-trained prediction models to accelerate materials design by integrating them into numerical optimization algorithms. 4. **Active Learning**: Offers sampling strategies based on Bayesian principles to balance exploration and exploitation, particularly useful in scenarios with limited data. 5. **Comprehensive Features**: Includes databases, outlier detection, and advanced ML algorithms for classification, regression, and surrogate optimization. The effectiveness of MLMD is demonstrated through various case studies, including perovskites, steel, and high-entropy alloys, showing its ability to enhance the efficiency and robustness of materials design. The platform aims to bridge the gap between material science and AI, facilitating the advancement of materials informatics and accelerating the discovery of advanced materials.MLMD is an AI platform designed to accelerate the discovery of advanced materials by providing a programming-free, user-friendly interface for materials design. The platform integrates various functionalities, including data analysis, descriptor refactoring, hyper-parameter optimization, and property prediction. It supports end-to-end materials design, from data collection and preprocessing to model inference and surrogate optimization, even in scenarios with limited data through active learning. MLMD's key features include: 1. **User-Friendly Interface**: No programming required, making it accessible to researchers without extensive technical backgrounds. 2. **End-to-End Materials Design**: Capable of discovering novel materials with desired properties through a seamless workflow. 3. **Surrogate Optimization**: Utilizes well-trained prediction models to accelerate materials design by integrating them into numerical optimization algorithms. 4. **Active Learning**: Offers sampling strategies based on Bayesian principles to balance exploration and exploitation, particularly useful in scenarios with limited data. 5. **Comprehensive Features**: Includes databases, outlier detection, and advanced ML algorithms for classification, regression, and surrogate optimization. The effectiveness of MLMD is demonstrated through various case studies, including perovskites, steel, and high-entropy alloys, showing its ability to enhance the efficiency and robustness of materials design. The platform aims to bridge the gap between material science and AI, facilitating the advancement of materials informatics and accelerating the discovery of advanced materials.