Key Topics

Machine Learning Algorithm

• Understand the difference between supervised, unsupervised, and reinforcement learning
• Learn common algorithms such as linear regression, decision trees, and neural networks
• Explore model evaluation techniques like cross-validation, precision/recall, and confusion matrices

Vectorization Algorithm

• Learn how to convert raw data (e.g., text, images, audio) into numerical vectors
• Apply methods like Bag-of-Words, TF-IDF, and Word Embeddings for NLP tasks
• Understand feature scaling, normalization, and dimensionality reduction (e.g., PCA)

Use cases of GAI

• Automate text generation for education, entertainment, and customer support
• Generate images, music, and videos for creative industries
• Enhance productivity with code generation, summarization, and workflow automation

Summary

We participated in a range of seminars given by Professor Thomas Miller which introduced us to core principles in modern AI. These included machine learning algorithms such as regression, decision trees, and neural networks as well as text vectorization methods including TF-IDF and word embeddings for Natural Language Processing. The seminars drew distinctions between rule-based AI from the past and current data-directed AI.

Through lab exercises, we practiced generating images from sentences describing stories using iterative refashioning of prompts, and then built reusable prompt templates that ensured quality, clarity and consistency in image generation. Finally, we applied data visualization methods to compare free-form versus template-based prompting, and reflected on how the use of templates can shape both creativity and outcomes.

Machine Learning Algorithms

Machine Learning (ML) equips systems with the ability to learn patterns from data and improve predictions over time. Instead of being programmed with explicit rules, ML algorithms discover structures in data and adapt to new information.