Data Science Pedagogy Winter Workshop
This is a hands-on Data Science pedagogy workshop for educators of underrepresented or underserved students in Data Science/Machine Learning.
Day 4 (Jan 7th): Neural Network Models for Classification
On this day, we will cover neural network models for classification. The materials will be broken down into a sequence of bite-sized concepts. Each concept will be introduced in a short 10-20 minute video; following each video, there will be a short concept-check quiz for the viewer to test their understanding. These materials are taken from DSC6232 Machine Learning and Computational Statistics, an intensive summer data science course run by IACS and the University of Rwanda. These materials serves as either an introduction of new data science concepts to participants or as a foundation for discussion amongst participants on how to present these concepts to their own students.
The workshop follows a flipped-classroom structure, so participants should watch the videos and review the materials before Jan 7th.
Morning Session
During the morning session of Day 4, participants will explore a Deepnote notebook that puts these concepts together in ways that expose interesting properties of these models.
Afternoon Session
During the afternoon session of Day 4, participants will discuss how to adapt existing materials or how to create new materials for teaching these data science concepts to their own students.
Topic 1: Neural Networks for Classification
Topic 2: Interpreting Neural Networks
Topic 3: Regularization for Neural Networks
Topic 4: Encoding and Transforming Data
Topic 5: Pre-Processing Data
Topic 6: Sources of Bias in Model Interpretation and Usage
-
Lecture Video:
(11:15 minutes) - Lecture Slides: Slides
- References:
- Dealing with Data Imbalance
- A Classification Procedure for Highly Imbalanced Class Sizes
- A Framework of Rebalancing Imbalanced Healthcare Data for Rare Event Classification
- Bias in Machine Learning: How Facial Recognition Models Show Signs of Facism, Sexism, and Ageism
- Machine Bias
- Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification
- Concept Quiz: Quiz
Topic 7: Variable Selection
Topic 8: Principal Component Analysis
Morning Session: Exploratory Notebooks
- Neural Networks for Classification
- Data Preprocessing and Model Interpretation Notebook
- PCA and Dimensionality Reduction