Machine Learning for Time Series — Master 2 MVA

Teaching material and outline of the course Machine Learning for Time Series (Master MVA) during 2025–2026.

Course description

In many application contexts (health, economy, advertising…), the data collected takes the form of time series. The fundamental challenge then consists in choosing an adapted representation, allowing to take into account the temporal information as well as possible.

Machine Learning for time series gathers a large number of unsupervised or supervised tasks such as prediction, classification, completion/interpolation, clustering, segmentation/change-point detection or anomaly detection. But in reality, most of the work for a data scientist dealing with temporal data consists in a series of hidden tasks:

Understand the data: know where they come from, how they were acquired, what are their characteristics, interact with domain-experts
Improve the data: find accurate representation spaces, consolidate the data (denoising, detrending, outlier removal)
Model the data: physical/statistical or expert-based models, simple, adaptive and interpretable models
Extract information from the data: find repetitive patterns, features of interest, change-points

This course aims to provide an overview of ML techniques to study time series, mostly focused on these often poorly-documented hidden tasks, widely illustrated on real data. Note that in its current form, the course will only marginally discuss Deep Learning algorithms.

Logistics

Lectures take place on Monday mornings at Université Paris Cité. On-site only — NOT filmed or recorded. Lectures in French, all material in English. Tutorial sessions: Monday mornings on-site OR Monday afternoons on Zoom (except Tutorial 3). Attendance mandatory. Tutorials led by Valerio Guerrini. Auditeurs libres cannot attend.

Planning

06/10/2025 09:00–12:00 Cordeliers · Amphi Frezal	Introduction
06/10/2025 09:00–12:00 Cordeliers · Amphi Frezal	Lecture 1 — Pattern Recognition and Detection Problem statement Comparing time series (Euclidean, DTW…) Detecting patterns in time series Learning patterns from time series
13/10/2025 09:00–12:00 Cordeliers · Amphi Frezal	Lecture 2 — Feature Extraction and Selection Feature extraction (statistical, spectral, symbolic, DL…) Feature selection (unsupervised & supervised)
27/10/2025 09:00–12:00 on-site OR 14:00–17:00 Zoom	Tutorial 1 — Lectures 1 & 2
03/11/2025 09:00–12:00 Cordeliers · Amphi Frezal	Lecture 3 — Models and Representation Learning Standard models (sinusoidal, trend+seasonality, AR, HMM) Representation learning (sparse coding, dictionary learning)
10/11/2025 09:00–12:00 Cordeliers · Amphi Frezal	Lecture 4 — Data Enhancement and Preprocessings Denoising (filtering, sparse/low-rank) Detrending Interpolation of missing samples Outlier removal
17/11/2025 09:00–12:00 Cochin · Amphi Luton	Lecture 5 — Change-Point and Anomaly Detection Change-point detection (cost functions, search methods, calibration) Anomaly detection (statistical, model-based, distance-based) Evaluation of event detection methods
24/11/2025 09:00–12:00 on-site OR 14:00–17:00 Zoom	Tutorial 2 — Lectures 3 & 4
01/12/2025 09:00–12:00 Cochin · Amphi Luton	Lecture 6 — Multivariate Time Series Models for multivariate time series (VAR, multivariate dict. learning) Graph signal processing (GFT, bandlimitedness, filtering, graph learning)
08/12/2025 09:00–12:00 Zoom OR 14:00–17:00 on-site	Tutorial 3 — Lectures 5 & 6
15/12, 17/12/2025 05/01, 07/01/2026 All day · Zoom	Oral presentations

Validation

Tutorials (25%): commented notebooks and/or PDF reports. Attendance at at least one session per tutorial (on-site or remote) is mandatory. Missed or late assignments → FAIL.
Mini-project (75%): one paper on a topic related to the course, done in pairs.
- Report (25%): PDF, 5 pages — template provided
- Source code (25%): commented Jupyter notebook
- Oral presentation (25%): 10 min with slides

Registration form distributed at the first lecture. Final registration deadline: 12 October 2025.

Additional resources

Useful references

List of possible topics / projects