Learn from Professors at EPFL
Taught on campus at EPFL, the DSFM program balances theory and practice to quickly cover the most-important aspects of Data Science. You will learn in a highly-interactive environment, through a combination of lectures, demonstrations, case studies, problems, and visualizations.
DSFM helps executives, managers, analysts, and other professionals understand when AI works, and when it does not. We also take time for brainstorming and networking with peers to help you convert opportunities into real-world results.
With backgrounds ranging from computational science to innovation strategy, DSFM professors and instructors are at the forefront of Data Science research and applications. This combination of expertise assures that your workshop will address all aspects of Data Science, keeping you at the cutting edge of emerging technologies and data thinking.
Teaching assistants are available throughout the day to provide one-on-one assistance with practical problems. All participants will leave the course being able to build, evaluate, and work with real data and real models!
Prof. Kenneth Younge is an Associate Professor at EPFL, the Chair of the Technology and Innovation Strategy Lab, and Program Director for DSFM. He has started four companies and worked in the roles of Director of Development, Consultant, CTO, and President.
Professor Younge currently teaches the Masters course on Data Science for Business, the Doctoral course on Computational Methods for Management, the IML course on Data Science for Logistics, and an eMBA course on Technology and Innovation Strategy. His research focuses on computational economics and digital transformation. His doctoral students and post-doctoral researchers collaborate with a wide range of Swiss and US companies in ongoing research projects.
We organize interactive lunch sessions during the DSFM Boot Camp. Additional professors and industry experts join the class to meet with small groups of DSFM participants to discuss particular topics of interest. Each discussion leader is an expert in a given area. Examples of discussion leaders include:
Prof. David Atienza is an Associate Professor at EPFL and expert on embedded systems for the Internet of Things (IoT). He leads a discussion on how smart wearables, wireless sensors, edge computing, and embedded machine learning work together to create new business opportunities.
Prof. Chris Tucci is the former Dean of the College of Management at EPFL and an expert on issues of design thinking and digital transformation. He leads a discussion on how firms construct data-driven strategies to transition to new technologies, business models, and organizational forms.
Prof. Dimitrios Kyritsis is an Adjunct Professor at EPFL and Director of the Doctoral Program on Robotics, Control and Intelligent Systems. He is an expert on the management of data and information flows, and D-I-K (Data-Information-Knowledge) transformations throughout the lifecycle of products.
Prof. Negar Kiyavash is a Full Professor at EPFL and Chair of Business Analytics in the College of Management. She is an expert on causal inference from networked big data and teaches on advanced topics in machine learning, artificial intelligence, optimization, and data science.
Prof. Bob West is an Assistant Professor at EPFL and head of the Data Science Lab in the School of Computer and Communication Sciences. His research aims to make sense of Big Data collected from the Web, such as server logs, social media, wikis, online news, online games, etc.
Prof. Jeffrey Kuhn is an Assistant Professor at the University of Carolina and a US patent attorney. He leads a discussion on how AI algorithms and proprietary methods can best be protected as either intellectual property or trade secrets.
Prof. Alex Biedermann is an Associate Professor at the UNIL Ecole des Science Criminelles and an expert on decision-making under uncertainty. He leads a discussion on how computational methods can support a more systematic approach for automating decisions.
Dr. Christopher Bruffaerts is a lecturer at the College of Management at EPFL and instructor for the Masters course on Data Science in Practice. He has worked on customer analytics, fraud detection, and big data technologies at BNP Paribas Fortis, Credit Suisse, and UPC.
The DSFM Boot Camp is held on campus at EPFL (the École Polytechnique Fédérale de Lausanne). EPFL is part of the Swiss Federal Institute of Technology and one of the leading technical centers in Europe.
EPFL is home to over 350 laboratories and research groups, each working at the forefront of science and technology – with a diverse, committed and stimulating research community that is active over a wide spectrum of quantitative and design-focused disciplines.
DSFM is not just for those living and working in Switzerland.
Lausanne is a destination in of itself - and worthy of a visit while you are here. Come for a course and stay for the weekend. Get to know one of Europe’s most vibrant and cosmopolitan centers for science and technology.
EPFL and Lausanne are less than an hour by train from the Geneva International Airport, and a short flight from most any city in Europe.
What programming skills are required for DSFM?
Please see the section on "Preparation" for each type of DSFM course. Some courses are more technical with more programming; others have no programming. Typically, when programming is involved, the objective is to learn "how to read code" so that you can ask good questions, get involved with your team, and understand the processing flow. Some Workshops, however, are gear more towards active development. Please check each course for more information or contact us with specific questions/concerns.
All of the examples and demonstrations will be given in Python. We therefore recommend that you review the basics of Python before the technical courses. A good place to start is the 7-hour Python tutorial by Kaggle. Doing so will help you follow along with the demos, examples, and project solutions covered in class. Of the two recommendations (to know some linear algebra and to be prepared with some basic Python), being prepared with some basic Python is more important.
What does "basic knowledge of linear algebra" mean?
DSFM courses cover machine learning models at a high level. To understand how/why a model works requires a deeper understanding of the optimization process by which each method minimizes error. The optimization process often is expressed in terms of linear algebra. It therefore is helpful if you have been exposed to linear algebra - at least through a beginning undergraduate level.
However, if you have forgotten your maths, then that is OK. One does not need to solve or prove any mathematical problems in DSFM. All of the methods that we cover (such as Gradient Descent) also will be described in terms of a visual representation of the process, and ultimately, one can still use such models without a complete understanding of the internal mechanics. Thus, the recommendation to have a "basic knowledge of linear algebra" is more of a suggestion, than a requirement..
Can you recommend an online Python course to help me prepare?
Yes! It is important to learn, refresh, or upgrade your Python skills before you arrive for the DSFM Boot Camp. Doing so will help you get much more out of the course. Below we list several options to help you get going:
The Blue Piste (7 hours)
The basic Kaggle course on Python: https://www.kaggle.com/learn/python
This is good for a fast overview, but it assumes that you already know some basic knowledge of coding.
If you do not know anything about coding (at all) then you should do the Red Piste.
The Red Piste (34 hours)
The complete JetBrains Academy track on Python: https://hi.hyperskill.org
This is a great way to go from absolute beginner to fully prepared. It is interactive and you learn in bite-sized exercises.
This track also includes powerful tools to help you track the concepts that youy you have studied: https://hyperskill.org/knowledge-map
The Black Piste (many more hours)
The comprehensive Kaggle course on Pandas: https://www.kaggle.com/learn/pandas
This is an advanced preparation in Python that has more of a focus on manipulating data.
The Black Piste is the most complete preparation for a DSFM course, especially for those who plan to return to work and immediately put their new skills into practice
I completed the "Blue Piste" (recommended above) - should I do more?
If you prepared the Blue Piste and it went well, then you'll be fine. The Red Piste starts out easier with less assumed prior knowledge, but by the end it provides an even more comprehensive foundation with more topics. If you completed the Blue Piste, but want to continue - then one option is to then bounce down the Red Piste, quickly, looking to pick up new topics.
But if you have the time, learning some of the Pandas functionality on the Black Piste will help participants who want to immediately return to work and quickly put what you learn into practice. .
Does DSFM teach participants how to manage Data Science teams?
The primary aim of DSFM is not about teaching the soft skills of "managing people" - DSFM is a technical course. The primary aim of DSFM is to give professionals (executives, managers, engineers, analysts, programmers, IT professionals, etc.) a fast, technical understanding of Data Science vocabulary and concepts.
Please keep in mind that there are two versions of DSFM: The Technical Boot Camp and the "No-Code" Fast Track.
The Fast Track version of DSFM skips over, and does not cover, the coding aspects of implementing Data Science models. The Fast Track does, however, cover the technical and conceptual basis of how each model works, how it can be evaluated, and how it can be improved.
The technical Boot Camp is more low-level in that it exposes participants to what is required to actually develop and implement Data Science solutions through programming code. The Boot Camp goes through many programming examples and exercises (about half of the overall time in the Boot Camp); but even so, the intention of the Boot Camp is not to turn participants into professional programmers, or get them to the point where they can develop production code on their own. Instead, the intention of the Boot Camp is to educate "technically capable" people (who must be computing-able, and not computing-phobic) to work in conjunction with data scientists or outside consultants.
DSFM participants may go back to work to actually "manage" data scientists - or the may not. There are many professionals in companies who will never code, but who nevertheless still need to understand the realistic strengths and limitations of Data Science and Machine Learning so that they can make better strategic decisions for their company.
Do you arrange transportation and lodging?
No - we ask you to arrange your own transportation and lodging. But when you enroll, we will send you instructions with more details about travel and several options for lodging nearby.
Will I be able to implement what I have learned at the end of the course?
For the Boot Camp, the course dedicates over 22 hours to programming demonstrations, exercises, and short programming projects. While most students approach the programming projects from a mindset of looking at the solutions and then backward engineering what it does, motivated students can also tackle each project with the intention of trying to program it in full - by yourself. If you do that on all the projects - then you will be ready to start implementing real models and solutions yourself at the end of the course. But that of course is up to you, your level of motivation, and so on. Almost every student leaves DSFM saying that the course is harder - and is more hands on - than they expected. In that sense, DSFM should put you on a path toward implementing real solutions.
I'm new to Data Science, will I be able to keep up?
DSFM provides an EPFL-level experience, but we recognize that you will be returning to the course from industry after many years out of school. We work hard to not to drop anyone during the course, and we have TAs on hand for 1-on-1 help.