Computational methods for cultural studies

This course is designed to equip students from humanities backgrounds with a practical understanding and hands-on experience in applying computational methods, particularly automated image classification, to cultural, intellectual, and visual historical research. By balancing fundamental Python programming with core AI and computer vision principles and culminating in applied project development, students will learn to connect theoretical concepts with practical implementation. The objectives focus on developing not only technical skills for data manipulation and model training but also critical thinking to interpret AI results within a historical and cultural context, fostering interdisciplinary insight.

-Python Fundamentals & Coding Proficiency: Acquire foundational Python programming skills, enabling them to write and understand basic programs. Utilize essential libraries (NumPy, Matplotlib) for effective data handling and visualization in a Colab environment.
-Understand AI & Image Data: Explain core AI/ML principles, differentiate learning paradigms, and describe image data structures for cultural research.
-Grasp Neural Network Basics: Describe how neural networks learn, including forward/backward propagation and gradient descent.
-Build & Train CNNs: Implement and train a basic Convolutional Neural Network (CNN) in PyTorch for image classification.
-Evaluate & Interpret Models: Analyze model performance using metrics like accuracy and confusion matrices and critically interpret AI results in a cultural context.
-Develop Applied Projects: Collaborate on a computational image classification project, connecting it to a cultural or historical research question.