Mediaspace scheduled maintenance: Aug 25, 2026 07:00 - 12:00 AM. During this time, videos will be temporarily unavailable. Check status updates.

Course

CS-421: Machine learning for behavioral data

Lectures in this course (26)

Structure Discovery: Tracing Student Knowledge

Introduces Bayesian Knowledge Tracing, Additive Factors Model, and clustering algorithms for tracing student knowledge and discovering structures.

Data Handling: Problems and Distributions

Covers common data problems and important distributions, along with correlation and dependencies analysis.

Regression and Classification

Explores regression, classification, linear models, decision trees, and random forests in data analysis.

Advanced Structure Discovery: Distance Metrics and Time Series Data

Explores clustering algorithms, distance metrics, and time series data analysis techniques.

Introduction to ML for Behavioral Data

Covers machine learning approaches for personalization and their real-world application, emphasizing homework, projects, and feedback.

Affective Computing: Theories, Applications, and Predictive Models

Explores emotion theories, applications, and predictive models in affective computing, analyzing NSF funding trends, emotion impact on education and medicine, and emotion detection through physiological signals and visual data.

Multimodal Analytics: Human-Robot Interaction and Learning Processes

Delves into Human-Robot Interaction, student engagement, and learning analytics using multimodal data to enhance educational processes.

Machine Learning for Behavioral Data

Introduces a course on Machine Learning for Behavioral Data at EPFL, covering ML algorithms, data handling, and model evaluation.

Data Exploration: Time Series Analysis

Explores data analysis techniques, emphasizing time series prediction in a flipped classroom setting.

Regression: Linear Models

Introduces linear regression, generalized linear models, and mixed-effect models for regression analysis.

Classification: Decision Trees and kNN

Introduces decision trees and k-nearest neighbors for classification tasks, exploring metrics like accuracy and AUC.

Model Evaluation

Delves into model evaluation, covering theory, training error, prediction error, resampling methods, and information criteria.

Knowledge Tracing

Delves into Bayesian Knowledge Tracing and Learning Curves, exploring the prediction of student knowledge over time and the importance of accurate performance measurement.

Deep Knowledge Tracing

Explores Deep Knowledge Tracing and its application in predicting student learning outcomes using neural networks and loss functions.

Deep Learning: Recurrent Neural Networks

Explores Recurrent Neural Networks for behavioral data, covering Deep Knowledge Tracing, LSTM, GRU networks, hyperparameter tuning, and time series prediction tasks.

Neuroscience: Behavioral Analysis

Explores neuroscience through behavior analysis, focusing on pose estimation and motor control studies.

Clustering Algorithms: K-Means vs Spectral Clustering

Compares K-Means and Spectral Clustering algorithms, highlighting their differences and practical applications in clustering student behaviors.

Time Series Clustering

Covers clustering time series data using dynamic time warping, string metrics, and Markov models.

Fairness in Machine Learning

Explores sources of unfairness in machine learning, the importance of fairness metrics, and evaluating model predictions using various fairness metrics.

Explainability: Methods and Applications

Introduces interpretability methods like PDP and LIME for understanding model predictions.