| (*)Students have a comprehensive understanding of supervised machine learning techniques as well as the ability to choose, implement, and adapt methods for classification, regression, and feature selection in scientific and engineering contexts. They can critically evaluate model performance using appropriate metrics and techniques, understand the mathematical principles behind supervised learning, and can apply these methods to complex data analysis tasks.
|
(*)- Implementing Classification and Regression Techniques (k4)
Students can apply foundational supervised learning methods to classify data and predict continuous targets, understanding the appropriate contexts for classification and regression.
- Evaluating and Interpreting Machine Learning Models (k5)
Students are able to evaluate model performance using metrics such as confusion matrices and ROC curves, and interpret the results to assess model accuracy, bias, and variance.
- Applying Model Validation and Hyperparameter Tuning (k4)
Students can perform cross-validation and optimize model hyperparameters, ensuring the robustness and generalizability of supervised machine learning models.
- Understanding and Implementing Advanced Algorithms (k4)
Students are capable of implementing advanced supervised learning algorithms, including logistic regression, support vector machines (SVMs), neural networks, and deep networks.
- Handling Complex Data Structures and Feature Engineering (k4)
Students can analyze sequence data (e.g., time series) and apply techniques like bagging, boosting, feature selection, and feature construction to enhance model performance.
- Balancing Bias and Variance for Model Optimization (k5)
Students are able to analyze and mitigate underfitting and overfitting in machine learning models by understanding and adjusting the trade-off between bias and variance.
|
(*)Students know of the fundamental principles and objectives of supervised machine learning, covering topics such as classification, regression, model evaluation, and feature engineering. They understand advanced algorithms and concepts like logistic regression, SVMs, neural networks, cross-validation, and techniques to manage model complexity, providing a basis for applying supervised learning to real-world data problems.
|
