DS-GA 3001: Course Schedule - Fall 2026

Weekly Schedule

⚠️ Work in progress: This schedule is tentative and is being actively updated ahead of and during the semester. Topics, readings, lab sessions, and due dates may change. Check course announcements and Discord for updates.

Lecture (DS-GA 3001.001) meets Mondays 4:55 PM - 6:55 PM. Lab (DS-GA 3001.002) meets Tuesdays 7:10 PM - 8:00 PM, the day after each lecture. Both sessions are at 60 Fifth Avenue, Room 150, unless otherwise noted.

September 7

Labor Day - No Class

Week 1: Lecture Sept 14 | Lab Sept 15

Introduction to Visualization for Machine Learning

Lecture Topics:

• Course overview and logistics
• What is visualization for machine learning?
• Information visualization fundamentals
• The machine learning pipeline and where visualization fits

Lab Session:

• Development environment setup
• Introduction to Vega-Lite
• Creating basic visualizations
• Git and project structure

Readings:

• Hohman et al. “Visual Analytics in Deep Learning: An Interrogative Survey” (2018)
• Munzner, “Visualization Analysis and Design”, Chapter 1

Assignment: TBD

Week 2: Lecture Sept 21 | Lab Sept 22

Perception and Color Theory for Visualization

Lecture Topics:

• Pre-attentive processing and visual channels
• Gestalt principles
• Color spaces and perceptual uniformity
• Sequential, diverging, and categorical scales
• Color accessibility

Lab Session:

• Perception and color in practice
• Building and evaluating color scales

Project: Team formation

Week 3: Lecture Sept 28 | Lab Sept 29

Model Assessment and Performance Metrics

Lecture Topics:

• Confusion matrices and ROC curves
• Visual analytics systems for model performance
• Calibration theory and practice

Lab Session:

• Model assessment in practice (materials in preparation)

Readings:

• Ren et al., “Squares: Supporting Interactive Performance Analysis for Multiclass Classifiers” (2016)
• Görtler et al., “Neo: Generalizing Confusion Matrix Visualization to Hierarchical and Multi-Output Labels” (2022)

Week 4: Lecture Oct 5 | Lab Oct 6

White-box Model Visualization

Lecture Topics:

• Linear regression and visual analytics systems
• Generalized Additive Models (GAMs) and Explainable Boosting Machines
• Tree-based models and visualization techniques
• Decision rules and global surrogate models

Lab Session:

• Interpretable ML methods

Readings:

• Mühlbacher & Piringer, “A Partition-Based Framework for Building and Validating Regression Models” (2013)
• Hohman et al., “Gamut: A Design Probe to Understand How Data Scientists Understand Machine Learning Models” (2019)

Due: Project proposal (4 pages)

October 12

Fall Break - No Class

Week 5: Lecture Oct 14 (Wednesday Make-up) | Lab Oct 13

Black-box Model Interpretation & Project Discussion

The University runs a Monday schedule on this date, replacing Fall Break. The Tuesday lab meets as usual on Oct 13 — the only week in which the lab precedes its lecture.

Lecture Topics:

• Partial Dependence Plots (PDP)
• Local Interpretable Model-agnostic Explanations (LIME)
• SHAP (SHapley Additive exPlanations)
• Project ideas and guidelines

Lab Session:

• Black-box explainability methods

Readings:

• Ribeiro et al., “Why Should I Trust You?” Explaining the Predictions of Any Classifier (2016)
• Molnar, “SHAP: A Unified Approach to Interpreting Model Predictions” (2024)

Week 6: Lecture Oct 19 | Lab Oct 20

Clustering Visualization

Lecture Topics:

• Introduction to unsupervised learning
• K-means clustering and DBSCAN
• Hierarchical clustering dendrograms
• Cluster validation and comparison techniques
• The manifold hypothesis and intrinsic dimensionality
• Default project overview

Lab Session:

• Interactive clustering interfaces
• Cluster exploration tools

Readings:

• Wolfram Clustering Tutorial - Required
• Shlens, “A Tutorial on Principal Component Analysis” (2014)

Week 7: Lecture Oct 26 | Lab Oct 27

Dimensionality Reduction

Lecture Topics:

• PCA and linear projections
• t-SNE: theory and pitfalls
• UMAP and modern techniques
• Topomap: topologically-constrained dimensionality reduction
• Projection quality metrics

Lab Session:

• Implementing projection views
• Interactive parameter tuning
• Projection comparisons

Readings:

• Wattenberg, Viégas & Johnson, “How to Use t-SNE Effectively” (2016) - Required
• van der Maaten & Hinton, “Visualizing Data using t-SNE” (2008)
• McInnes et al., “UMAP: Uniform Manifold Approximation and Projection” (2018)

Due: Mid-term project update (1 page)

Week 8: Lecture Nov 2 | Lab Nov 3

Deep Learning Visualization

Lecture Topics:

• Deep learning terminology and foundations
• Linear models and loss functions
• Shallow and deep neural networks, activation functions
• Neural network architecture visualization
• Activation and gradient visualization
• CNN filter visualization

Lab Session:

• TensorBoard integration
• Building network diagrams
• Interactive layer exploration

Readings:

• Prince, “Understanding Deep Learning” (2023), Chapters 2-4
• Zeiler & Fergus, “Visualizing and Understanding Convolutional Networks”
• Olah et al., “The Building Blocks of Interpretability”

Week 9: Lecture Nov 9 | Lab Nov 10

NLP and Large Language Model Visualization

Lecture Topics:

• NLP basics and general text visualization
• Word embeddings visualization
• Attention in transformers
• Examples of RNN and LLM visualization

Lab Session:

• Building text visualization systems
• Interactive embedding exploration

Readings:

• Vaswani et al., “Attention is All You Need” (2017) - Foundational
• Vig, “BertViz: A Tool for Visualizing Multi-Head Self-Attention” (2019)
• Strobelt et al., “LSTMVis: A Tool for Visual Analysis of Hidden State Dynamics in RNNs” (2017)

Week 10: Lecture Nov 16 | Lab Nov 17

Topological Data Analysis

Lecture Topics:

• Introduction to topology and Betti numbers
• Persistence diagrams and homology
• Simplicial complexes and Vietoris-Rips construction
• The Mapper algorithm
• Applications in biology, chemistry, and machine learning

Lab Session:

• TDA tool integration (GUDHI, scikit-tda, Ripser, KeplerMapper)
• Interactive topology exploration

Readings:

• Chazal & Michel, “An Introduction to Topological Data Analysis” (2021) - Required
• Carlsson, “Topology and Data”

Week 11: Lecture Nov 23 | Lab Nov 24

Time Series and Streaming Data

Lecture Topics:

• Temporal model performance
• Concept drift visualization
• Real-time monitoring
• Anomaly detection displays

Lab Session:

• Streaming visualization implementation
• Dashboard design patterns

Readings:

• Selected papers on temporal visualization

November 26-27

Thanksgiving Recess - No Class

Note: Thanksgiving recess falls on Thursday and Friday and does not affect our Monday lecture or Tuesday lab.

Week 12: Lecture Nov 30 | Lab Dec 1

Interpretable ML and Fairness

Lecture Topics:

• Fairness metrics visualization
• Bias detection interfaces
• Interpretability vs accuracy
• Ethical considerations

Lab Session:

• Fairness dashboard creation
• What-if tool exploration

Readings:

• Selected papers on ML fairness

Week 13: Lecture Dec 7 | Lab Dec 8

Project Presentations I

• Group project presentations (Part 1)
• Peer feedback sessions
• Q&A and discussions

Week 14: Lecture Dec 14 (no lab)

Project Presentations II

• Group project presentations (Part 2)
• Course wrap-up
• Future directions in VisML

Project: Final reports due December 14

Important Notes

1. No Class: September 7 (Labor Day) and October 12 (Fall Break)
2. Make-up Lecture: Wednesday, October 14 - the University runs a Monday schedule that day
3. Last Lab: Tuesday, December 8 - Week 14 (Dec 14) has no lab session
4. Office Hours: TBA
5. Slides: Available after each class on course website
6. Recordings: Posted for registered students who miss class

Assignment Summary