Skip to main content
About HEC About HEC
Summer School Summer School
Faculty & Research Faculty & Research
Master’s programs Master’s programs
Bachelor Programs Bachelor Programs
MBA Programs MBA Programs
PhD Program PhD Program
Executive Education Executive Education
HEC Online HEC Online
About HEC
Overview Overview
Who
We Are
Who
We Are
Egalité des chances Egalité des chances
HEC Talents HEC Talents
International International
Sustainability Sustainability
Diversity
& Inclusion
Diversity
& Inclusion
The HEC
Foundation
The HEC
Foundation
Campus life Campus life
Activity Reports Activity Reports
Summer School
Youth Programs Youth Programs
Summer programs Summer programs
Online Programs Online Programs
Faculty & Research
Overview Overview
Faculty Directory Faculty Directory
Departments Departments
Centers Centers
Chairs Chairs
Grants Grants
Knowledge@HEC Knowledge@HEC
Master’s programs
Master in
Management
Master in
Management
Master's
Programs
Master's
Programs
Double Degree
Programs
Double Degree
Programs
Bachelor
Programs
Bachelor
Programs
Summer
Programs
Summer
Programs
Exchange
students
Exchange
students
Student
Life
Student
Life
Our
Difference
Our
Difference
Bachelor Programs
Overview Overview
Course content Course content
Admissions Admissions
Fees and Financing Fees and Financing
MBA Programs
MBA MBA
Executive MBA Executive MBA
TRIUM EMBA TRIUM EMBA
PhD Program
Overview Overview
HEC Difference HEC Difference
Program details Program details
Research areas Research areas
HEC Community HEC Community
Placement Placement
Job Market Job Market
Admissions Admissions
Financing Financing
FAQ FAQ
Executive Education
Home Home
About us About us
Management topics Management topics
Open Programs Open Programs
Custom Programs Custom Programs
Events/News Events/News
Contacts Contacts
HEC Online
Overview Overview
Executive programs Executive programs
MOOCs MOOCs
Summer Programs Summer Programs
Youth programs Youth programs
Instant

Using deep learning to predict the unpredictable

Economics
Published on:

Predicting outcomes as diverse as electricity consumption, air quality, subway ridership and market fluctuations is now possible thanks to deep learning. Malo Huard explains this new tool.

Using deep learning  to predict the unpredictable - ©madgooch-AdobeStock

Machine learning , which is based on learning by experience, can be harnessed to solve an array of problems, with applications including object recognition (faces, diagrams, writing, and so on); diagnostic support; detection of credit card fraud; stock market analysis; and DNA sequence classification. To produce reliable results, however, the approach requires a vast amount of data, which is lacking in some fields.

"When I went to work on the predictive maintenance of railway switches for SNCF," explains Malo Huard, "I only saw a single failure out of 80,000 items of data I observed." What could Huard do in such a case, when the data was lacking? He decided to find a solution by thinking about the problem the other way around, by trying instead to detect the parameters of normal operations. By modeling all the signals through observation, the researcher was able to identify the precursors of a breakdown. "With this model,” says Huard, “we even know when teams on the ground are lubricating the point mechanics, because of the anomalous signals that we receive during that process.”

In the face of difficult choices: combining the best models

Like many other engineers and researchers attempting to make predictions, Huard uses mathematical models to obtain his results. While most choose models for their reliability in the context of the problem they are endeavoring to solve or to combine several models uniformly, however, Huard — in tandem with his thesis supervisor, Gilles Stoltz — opted for a different approach: weighting the models based on data history.

"Using this technique is almost as good as using the best model, even if you don’t know which one to choose or don’t know in advance," Huard says. This solution avoids randomly selecting a model to instead take advantage of the panoply of different methods. Huard’s approach won him second place — out of 479 participants — in the RTE challenge on forecasting electricity consumption . The aim of the competition was to predict how much energy the national and regional grids would use over 10 days (the duration of the challenge), after one day and per quarter-hour. Huard was the only competitor to aggregate the models sequentially.

 

Using this technique is almost as good as using the best model, even if you don’t know which one to choose or don’t know in advance.

 

Introducing competition between models and developing automatic learning

How does aggregation work? “I employ a range of automatic learning techniques to build elementary models, including neural networks,”  Huard explains. “The next step involves devising a meta-prediction system using sequential aggregation algorithms." At the start of the process, all the elementary models are aggregated uniformly (with the same weight). The weighting is then updated over time according to the data history. The model is particularly robust since it can work with very limited data: three months of data are enough to calibrate a model, for example.

This robustness is linked to the sequential aggregation, which sets the models in competition and deactivates those that are only reliable over a short timespan and cannot be widely applied . Aggregating models has further advantages: it is very inexpensive because the algorithms, which operate via updating, only sweep the data once compared to hundreds of times in the case of conventional machine learning programs. They are also faster. To achieve reliable results, it remains necessary to use good models in the aggregation. 

From predicting sales to air quality: a whole host of applications

The above method can be used in many different contexts. For example, Huard applies the techniques of sequential aggregation to forecasting product sales for Cdiscount — a crucial operation for logistics coordinators who have to anticipate demand and build up stocks in advance. As well as being involved in predicting electricity consumption, Huard is also interested in predictive maintenance. He is currently working on a project with EDF looking at the timetables for activating dams based on electricity demand.

The goal is to automate tasks currently carried out manually. All temporal processes can benefit from this approach, since few other models offer as reliable guarantees: forecasting air quality (realized by Stoltz), footfall in subways or airports, Vélib usage, stock prices, crude oil production, fluctuations in exchange rates, and so on. The fact that Huard’s laboratory works so closely with business means that the researcher is immersed in environments where he can explore and implement sequential aggregation applications, thereby helping industry gain the most from these solutions.

Based on an interview with Malo Huard, Ecole Polytechnique graduate and PhD researcher. Malo Huard, a graduate of the Ecole Polytechnique (2015), is working on a thesis in the Hadamard PhD School of Mathematics at Université Paris Saclay as part of the HEC Paris Management Research Group (CNRS joint laboratory / HEC Paris : GREGHEC). Huard’s dissertation is based on the robust aggregation of predictors. The methods that he has developed enable automatic predictions to be made using fundamental predictors (i.e., without the need for control).

Related content on Economics

Subscribe button for Knowledhe@HEC newsletter

Newsletter knowledge

A monthly brief in your email box and 3 issues of the book per year.

follow us

Insights @HECParis School of #Management

Follow Us

Support Research

Our articles are produced thanks to our reader's support