Banz 2025 – Overview

Welcome to the Machine Learning for Interdisciplinary Audiences course — a modular, discussion-driven program designed to introduce core machine learning (ML) concepts, practical tools, and societal implications to participants from diverse academic backgrounds.

Whether you’re coming from the natural sciences, social sciences, or the humanities, this course is designed to be accessible, thought-provoking, and hands-on.

👉 Jump directly to: Technical Topics | Societal Topics

🧭 Course Structure

The course consists of five interconnected components that blend foundational knowledge, interactive practice, and critical reflection:

📣 1. Introductory Talk

We begin with a shared foundation: What is machine learning, and how does it work in practice?

This opening session creates a common language for all participants — no prior knowledge assumed. It introduces key ideas and terminology, including:

  • What machine learning is, and how it differs from traditional programming
  • Types of learning (e.g. supervised vs. unsupervised)
  • The structure of an ML project (data, training, evaluation)
  • Conceptual challenges like overfitting, generalization, and interpretability

The session ensures everyone — regardless of background — is equipped to engage meaningfully in both the technical and societal parts of the course.

🧠 2. Technical Sessions

Over four 75-minute sessions, we explore core machine learning paradigms that power real-world applications:

  • Classification — assigning categories to data
  • Regression — predicting numerical values
  • Tree-based models — interpretable and widely used methods
  • Neural networks — flexible models inspired by the brain

Each session includes a brief instructor introduction, participant mini-presentations, and a structured discussion. The focus is on building intuition and insight — not mathematical depth — so that all participants can confidently interpret and apply ML techniques in their fields.

➡️ See the full Technical Topics Overview

💻 3. Python for Data Science Workshop

This session offers a fast-paced, beginner-friendly introduction to Python for data science and machine learning. No prior programming experience is required.

Working in browser-based Jupyter notebooks, participants learn to:

  • Use Python for basic logic and calculations
  • Manipulate structured data with pandas
  • Visualize results using matplotlib
  • Train simple models with scikit-learn

We end with a capstone project analyzing real weather data across multiple cities — integrating the full data science workflow from input to insight.

➡️ Browse the Python course on GitHub

🔎 4. Hands-On ML Sessions

Theory meets practice in these 45-minute interactive sessions. Participants use real code to explore how machine learning models behave — from visualizing decision boundaries to experimenting with training processes.

All materials run directly in the browser via Google Colab — no installation needed.

These sessions aim to:

  • Reinforce ML concepts through guided exploration
  • Encourage curiosity and intuition
  • Make model behavior tangible and transparent

Topics include:

  • Linear regression and gradient descent
  • Decision trees and random forests
  • Neural networks and loss landscapes

➡️ Explore the notebooks on GitHub

🌍 5. Societal Topic Sessions

Finally, we zoom out and ask: What are the real-world consequences of AI?

In four 60-minute sessions, participants explore the ethical, legal, and societal dimensions of artificial intelligence. These are participant-led, discussion-based conversations that draw on the group’s disciplinary diversity.

Core themes include:

  • 🧭 The EU AI Act and global regulatory trends
  • ⚖️ Algorithmic bias and fairness in decision-making
  • 🏥 AI in healthcare and diagnostics — promises and pitfalls
  • 📣 Misinformation, automation, and the future of work

Rather than seeking definitive answers, these sessions aim to sharpen critical thinking and connect technical understanding with societal impact.

➡️ See the full Societal Topics Overview

This course has been field-tested in interdisciplinary settings — including university workshops and summer schools — with participants from physics, law, philosophy, economics, biology, and beyond. It’s designed to be inclusive, flexible, and deeply engaging — regardless of prior experience.