Machine Learning for Time Series (Master MVA)
Teaching material and outline of the course Machine Learning for Time Series (Master MVA) during 20232024.
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, query by content/indexation, clustering, segmentation/changepoint detection or anomaly detection. But in reality, most of work for a data scientist dealing with temporal data consists in a series of hidden tasks such as:
 Understand the data: know where they come from, how they were acquired, what are their characteristics, interact with domainexperts and understand their problems
 Improve the data: find accurate representation spaces where the events of interest can be seen, consolidate the data (denoising, detrending, detection/removal of outliers)
 Model the data: physical/statistical or expertbased models, simple, adaptive and interpretable models
 Extract information from the data: find repetitive patterns, features of interest, changepoints
This course aims to provide an overview of ML techniques to study time series, in different tasks such as pattern extraction and recognition, anomaly detection, prediction, interpolation etc. The course will mostly focus on these often poorlydocumented hidden tasks and introduce several recent ML methods that will help the future data scientist to mine, but above all to understand time series. The course will be widely illustrated on real data and problems from current challenges and will emphasize aspects related to data understanding and interpretation. Note that in its current form, the course will only marginally discuss Deep Learning algorithms.
Outline and planning
Lectures will take place on Thursday afternoons at ENS Paris Saclay. Lectures will be onsite (ENS Paris Saclay) and will NOT be filmed or recorded. Lectures will be in French but all material (slides, homeworks...) is in English. For the tutorial sessions two options will be available: Thursday mornings will be remote on Zoom and Thursday afternoons will be onsite at ENS Paris Saclay. Attendance to the lectures and tutorial sessions is mandatory.
05/10/2023 14:00 → 17:00 Amphi Hodgkins (0I10) 
Introduction


Lecture 1: Pattern Recognition and Detection
 Problem statement
 Comparing time series
 Euclidean distance
 Normalized Euclidean distance
 Dynamic Time Warping (DTW)
 Detecting patterns in time series
 Euclidean distance
 DTW
 Learning patterns from time series
 Distancebased pattern extraction
 Dictionarybased pattern extraction


12/10/2023 14:00 → 17:00 Amphi Hodgkins (0I10) 
Lecture 2: Feature Extraction and Selection
 Problem statement
 Feature extraction
 Stationarity and ergodicity
 Statistical features
 Spectral features
 Local symbolic features
 Information theory features
 Deep Learning features
 Other features
 Feature selection
 Unsupervised setting
 Supervised setting


19/10/2023 09:00 → 12:00 Zoom OR 19/10/2023 14:00 → 17:00 Amphi Hodgkins (0I10)

Tutorial 1 on Lectures 1 & 2


26/10/2023 14:00 → 17:00 Amphi Hodgkins (0I10) 
Lecture 3: Models and Representation Learning
 Problem statement
 Standard models
 Sinusoidal model
 Trend+Seasonality model
 AR models (and variants)
 Hidden Markov model
 Representation learning
 Standard representations
 Notion of sparsity
 Sparse coding
 Dictionary learning


09/11/2023 14:00 → 17:00 Amphi Hodgkins (0I10) 
Lecture 4: Data Enhancement and Preprocessings
 Problem statement
 Denoising
 Filtering
 Sparse approximations
 Lowrank approximations
 Other techniques
 Detrending
 LeastSquare regression
 Other techniques
 Interpolation of missing samples
 Polynomial interpolation
 Lowrank interpolation
 Modelbased interpolation
 Outlier removal
 Isolated samples
 Contiguous samples


16/11/2023 09:00 → 12:00 Zoom OR 16/11/2023 14:00 → 17:00 Amphi Hodgkins (0I10)

Tutorial 2 on Lectures 3 & 4


23/11/2023 14:00 → 17:00 1Z14 
Lecture 5: ChangePoint and Anomaly Detection
 Problem statement
 Change point detection
 Dealing with nonstationary time series
 Problem statement
 Cost functions
 Search methods
 Finding the number of change points
 Anomaly detection
 Outlier detection
 Statistical methods
 Modelbased methods
 Distancebased methods
 Evaluation of event detection methods


30/11/2023 14:00 → 17:00 Amphi Hodgkins (0I10) 
Lecture 6: Multivariate Time Series
 Problem statement
 Models for multivariate time series
 Vector autoregressive models
 Multivariate dictionary learning
 Graph signal processing
 Concepts and definitions
 Graph Fourier Transform (GFT)
 Bandlimitedness and smoothness
 Graph filtering
 Graph learning


07/12/2023 09:00 → 12:00 Zoom OR 07/12/2023 14:00 → 17:00 Amphi Hodgkins (0I10)

Tutorial 3 on Lectures 5 & 6


20/12/2023 (all day) 22/12/2023 (afternoon) 11/01/2024 (all day) 12/01/2024 (all day) Zoom 
Oral presentations


Registration and mailing list
A registration form will be sent to all MVA students to subscribe to the course mailinglist. The final registration date is set to October 15th 2023 : no registration will be allowed after this date.
Tutorials
Tutorials will consist in interactive sessions where the students will be introduced to useful Python packages for time series analysis and have the opportunity to use and apply the different algorithms studied during the lectures. After each tutorial, students will be asked to work in pairs on a small project that will consist in (almost) direct applications of the algorithms seen in the tutorial sessions. Students are required to bring their own personal computer during the tutorial sessions : details on installation and required configuration will be provided before the first tutorial session. Attendance at at least one of the two tutorials sessions (onsite or remote) is mandatory: absences must be justified, otherwise you will receive a FAIL. Missed or late assignments will also give you a FAIL in the course.
Validation
 Tutorials (25%): commented notebooks and/or PDF reports
 Miniproject (75%): Choice of one paper on a topic related to the course. A list of possible topics/projects will be provided, but students can bring their own topics. In this case, they must contact the lecturer in advance for approval. Miniprojects will be done in pairs.
 Report (25%): PDF file, 5 pages  a template will be provided
 Source code (25%): commented Jupyter notebook
 Oral presentation (25%): 10 min presentation with slides
Additional ressources
Useful references


List of possible topics/projects

