News Categories: Bildung

Despite their 20-year age difference, Johan Rochel and Baptiste Lecoeur look similar. And given their different backgrounds, their paths were never really meant to cross. Yet they found themselves working together in the fall 2023 semester to design and teach a new Master’s-level class, which was given to students in several sections of EPFL’s School of Computer and Communication Sciences.

Rochel holds a PhD in law and philosophy and is a lecturer at EPFL’s College of Humanities (CDH). He’s also a co-founder of Ethix, a Zurich-based consulting firm in innovation ethics and the law. Meanwhile, Lecoeur is a second-year Master’s student in data science and a co-president of AGEPoly, EPFL’s general student association.

Rochel and Lecoeur form one of around 15 teaching tandems set up at EPFL over the past few months, on subjects ranging from materials science and physics to energy & the climate – another new class at EPFL. The idea with these tandems is to support teachers who want to introduce more sustainability-oriented material into their classes and to give students an opportunity to make a tangible contribution to EPFL’s shift in teaching approach while earning a little money. “This approach to education could also be useful for topics other than sustainability,” says Jacopo Grazioli, a project officer at EPFL’s Sustainability Unit, which initiated the tandems.

How does sustainability fit in with Rochel and Lecoeur’s new Master’s class on ethics in artificial intelligence? Rochel explains: “Ethics touches on all three aspects of sustainability: the economy, our society and the environment. In fact, only one of our lectures was devoted entirely to how AI affects power consumption and resource depletion. But as soon as you talk about the kinds of data used to train AI algorithms, or about design, data processing and the technology transition more broadly, you’re also talking about sustainability.”

We’d like AI to help us address certain aspects of the economy, for example, but that has an environmental and societal cost. So how do we strike the right balance?

“Of course, there’s also an environmental aspect to all the issues associated with justice, labor conditions, social interaction, the digital divide, and the inequality resulting from that divide” he adds. “What’s interesting from an ethics standpoint is when you have to make trade-offs between different priorities. For instance, we’d like AI to help us address certain aspects of the economy, for example, but that has an environmental and societal cost. So how do we strike the right balance?”

Worried about the future

For Lecoeur, getting involved in sustainability was a no-brainer. “Most of us students are worried about the future, and we generally have an open mind on societal issues,” he says. “That makes sense when you’re a scientist or engineer, since you develop methods and systems that have an impact on society and the environment.”

“That said, our concerns go beyond the environment,” he continues. “For now AI is largely unregulated, and it takes a lot longer for new laws to be passed than for new technology to be developed. Yet despite the lack of regulation, some AI programs are starting to be used on a massive scale.”

Lecoeur believes that one benefit of his Master’s class is that it raises other issues. “I think people at EPFL don’t always seek to understand the world around them beyond just the science. But our class is multi-disciplinary, at the crossroads of science and society.”

Rochel agrees. “I thought students knew more about the political systems in Switzerland and the EU, such as the separation of powers. In 2024 I’ll spend more time describing the government institutions that are at the center of efforts to enact regulations.”

“A good test”

Rochel and Lecoeur are both clearly pleased with being paired up. “We’ve been meeting every week since the class started in September to get our slides ready,” says Rochel. “I used to make a first draft and then Baptiste would comment on it and suggest improvements on both the content and the layout, looking for new ideas and materials. Showing the slides to Baptiste was a good test, since he would tell me if they were too heavy on theory or needed more examples.”

For his part, Lecoeur liked the class format, which differs from the one typically used in data science. “What’s great is that students work on the case studies in small groups. They get into animated discussions – it’s the first time I’ve seen that in my four years at EPFL. It’s crazy that we never sit down in groups to talk about things other than purely technical subjects. This is one way our class adds a lot of value, since it’s not something students get to do elsewhere.”

Students work on the case studies in small groups. They get into animated discussions – it’s the first time I’ve seen that in my four years at EPFL.

“People say EPFL is great for teaching problem-solving, but not so much for the soft skills – management, relating with colleagues, and so on,” says Lecoeur. “The students themselves have to fill in the gaps, and it’s great we can help them do that in our class.”

Word of mouth

Rochel and Lecoeur’s partnership ended at the end of the semester, but it’ll leave a lasting mark: their joint efforts have made the class more impactful and helped to hone the teaching materials – a real advantage for future years. They believe that by working together they have been able to get better feedback from students, who are more likely to open up to their peers.

“It’s such a great way to design a class,” says Lecoeur. “The thing is, students can be pretty unforgiving. When a new class doesn’t run smoothly from the outset, they spread the word quickly and few people take it. But if you’re able to get student input and come up with a format that works, then they’ll tell people how awesome it is – and more people will sign up as a result. Nothing’s more effective in drumming up interest than word of mouth!”

On this occasion, twenty-three teachers from eleven schools in the Canton of Vaud were presented with their certificates by EPFL President Martin Vetterli, in the presence of Frédéric Borloz, Head of the Département de l’enseignement et de la formation professionnelle (DEF), Professor Francesco Mondada, Academic Director of the Center LEARN, and Yann Secq, Executive Manager of the CAS ESIS.

Martin Vetterli opened the ceremony with a reminder that, with the rapid development of artificial intelligence, it was crucial to acquire a thorough understanding of these technologies to prevent them from taking on a magical aspect. „Magic isn’t quite what we ought to be teaching at school. What we need to teach in schools is to understand things, and in order to understand ChatGPT, it helps enormously to have been exposed to computational thinking, to know what an algorithm is, or to know what a probability is,“ he said.

A tailored program

This program, developed in response to strong demand for training for secondary school teachers as part of a trial phase for a new period of Computer Science in the Cycle 3 curriculum in the canton of Vaud, took place over 20 days. Delivered at EPFL by the Center LEARN team, it was accompanied by weekly hands-on practice for two years in the participants‘ respective classrooms.

An important feature of this continuing education program was that the pedagogical activities for students were developed entirely in collaboration with the participants. Teachers from a variety of educational backgrounds, including mathematics, arts and crafts, history, French and physical education, were thus able to appropriate the fundamental concepts of computer science and actively contribute to the development of relevant and engaging teaching sequences for their classes.

„The exceptional quality of the work produced by our participants is a testament to the extraordinary commitment and dedication they have shown over the past two years,“ said Yann Secq, Executive Manager of CAS ESIS.

Active teaching and concrete projects

Another special feature of these activities is their „learning by doing“ approach, which progressively takes the form of unplugged (computer-free) sequences, such as a board game called „Les Cordées du Cervin“, plugged-in activities involving the programming of small video games and interactive scenarios, before culminating in projects based on tangible objects, notably with the Thymio robot and the micro:bit minicomputer.

Commending the commitment of all the participants, Mr. Borloz emphasized: „The best gift we can give our children is to have knowledge of the digital world, rather than being subjected to it. This graduation is not a trivial event. For the students who leave school tomorrow, this knowledge opens the way to professional and academic training, while arousing their curiosity about MINT-related professions“.

This pioneering cohort is now giving way to a second group, which began in June 2022 and this time composed of eighteen teachers who will complete their course next July.

A team consisting of researchers and experts in digital education, including Dr. Sunny Avry, Prof. Francesco Mondada, Grégory Liégois, Elliot Vaucher, Felipe Martinez, and Dr. Jessica Dehler Zufferey, delved into the use of ChatGPT by teachers in public high schools and private institutions in the canton of Vaud. Conducted between February 22nd and March 22nd, 2023, in collaboration with the Direction Générale de l’Enseignement Postobligatoire and the Association Vaudoise des écoles privées, this pioneering survey in the region gathered insights from 931 teachers in public high schools and private schools in the canton.

The results reveal that 98% of the surveyed teachers are familiar with ChatGPT, and 52% of them have already used it. Among those who have used ChatGPT, 24% have done so as part of their teaching, primarily for lesson preparation. On the other hand, among the teachers who have not used it, reasons cited include a lack of time, interest, or need, as well as concerns about personal data sharing with OpenAI, the profit-oriented company behind ChatGPT.

„Teaching is primarily a profession based on relationships, and ChatGPT cannot replace the human relationship between teacher and student.“

Faced with the introduction of this tool, teachers experience primarily an increased sense of responsibility (56%), followed by fear (43%) and awe (34%). While ChatGPT impresses with its ability to perform certain tasks, it is important for teachers to consider ways to mitigate its potential negative impacts on education.

For example, most teachers believe that ChatGPT can facilitate cheating (87%), which calls for a review of evaluation methods (78%). They also believe that ChatGPT will not assist them in their work (74%) and does not promote student learning (55%). In the comments section of the survey, one teacher stated, „Teaching is primarily a profession based on relationships, and ChatGPT cannot replace the human relationship between teacher and student.“

Seeking Information Rather Than Prohibition

However, the idea of banning the tool in schools is rejected by 76% of the surveyed teachers. Another teacher expressed, „It is neither possible nor desirable to completely ban ChatGPT in schools. Such a prohibition could deepen social inequalities and encourage students to circumvent the rules.“ Moreover, over half of the teachers believe that students should be taught how to use ChatGPT in schools.

In parallel, 76% of the teachers expressed a desire to receive information and training on ChatGPT, particularly regarding its pedagogical use. As for the preferred formats for these training sessions, teachers mostly opt for PDF documentation, video tutorials, and in-person training.

The findings of this study have the potential to serve as a basis for guiding future decisions and initiatives regarding the training and support of teachers in the use of AI, enabling them to navigate an ever-evolving digital environment and make the most of these new tools.

The Center LEARN at EPFL has already collaborated with various institutions in this regard and is available for the development of research projects and training programs aimed at supporting teachers in this digital transition.

Over the last decade, the development of artificial intelligence has accelerated exponentially. Nowadays, the scope of Machine Learning applications has expanded to include almost every sector and industry, taking on a crucial role in the fields of science and engineering.

While their added value is no longer in question, unforeseen pitfalls can hide behind this promising technology. Society, democracy, and the environment are all spheres that can be compromised by the ethical drifts that algorithms can potentially cause, particularly when biases and their consequences are not considered.

At EPFL, no less than 223 courses address concepts related to Machine Learning, and more and more teachers are working to integrate ethical considerations in their lessons. However, these are by nature difficult to address and require the development of specific teaching approaches.

Cécile Hardebolle is a researcher who specializes in engineering learning and teaching. Together with Patrick Jermann and Maria Carla Di Vincenzo, she is leading the development of a new pedagogical tool which takes the form of a game based on an interactive story. Drawing from case studies, this tool allows students to virtually confront and explore these issues in a safe environment.

Learning by playing

The game was originally a prototype developed by two Master’s students, Alexandre Pinazza and Ester Simkova, as part of the course „How People Learn: Designing Learning Tools II“ taught by Roland Tormey.

„The goal of the exercise was to design a tool to support and integrate the learning of concepts related to ethics and sustainability in the fields of science and engineering. We really liked the idea proposed by Alexandre and Ester, which is based on an interactive scenario, a bit like those choose-your-own-adventure books. So we refined their idea by reworking the narrative structure to include analysis, reflection, and self-evaluation of affective reactions,“ explains Cécile Hardebolle.

After a laboratory evaluation phase (which results are currently being published) and a first pilot in the „Machine Learning for Behavioral Data“ class of Tanja Käser last spring, the game was tested by teachers in engineering schools during the 50th annual conference of the European Society for Engineering Education (SEFI).

The goal this semester was to validate the integration of the game in a classroom. „The framework of Nicolas Flammarion and Martin Jaggi’s course was perfect for this because they are both interested in digital ethics issues and had already started introducing these concepts into their program last year,“ says Hardebolle.

The device was thus debuted in the course „CS-433: Machine Learning,“ taught to more than 500 students by the two professors.

Collaborating to address a shared concern

„We might think that an algorithm is not a human reasoning and therefore it will be impartial and won’t have biases in its decision-making, but this is completely false. What is important to keep in mind is that the data we use contains a lot of bias. The entire process from data collection, its usage, and decision-making by algorithms often not only preserves these biases but also introduces new ones.

Ultimately, we reach a decision that can have significant consequences, for example, for an underrepresented group,“ explains Nicolas Flammarion. „Our goal was mainly to train engineers who have a vocation to take responsibility and make decisions in academia as well as in the private sector in companies.“

This observation echoes with Cécile Hardebolle’s viewpoint: „It is really important that we train our students to reflect on the negative impact that the technology they are going to design can have. How can we ensure that the expected benefits are not completely ruined by the harm that this technology may cause?“

The objective is to develop our engineers‘ ability to respond to the needs of society while minimizing the risks associated with technology development

A virtual experience

The students had a week prior to Nicolas Flammarion’s class to play online. The game immerses students into the role of a data scientist mandated to develop models that raise important ethical considerations.

Before a design decision, the scenario pauses and asks the students to reflect and justify their choices. The story then continues based on the consequences generated by these choices confronting the future engineers with their own cognitive biases and the dangers related to the nature of the data used. At the end, the game proposes to re-examine the choices that led to negative impacts and provides students with the opportunity to reformulate them, putting emphasis on emotions they felt.

A debriefing session links the game experience with its underlying ethical concepts, then the course provides mathematical approaches to identify, evaluate, and quantify biases and reviews existing methods for reducing them.

However, equity criteria are subject to limits and do not eliminate all issues. „It is important to keep in mind that Machine Learning has no guarantee of being aligned with our societal values if we do not take necessary measures,“ emphasizes Nicolas Flammarion.

„This is why it seemed necessary to me to develop a dedicated chapter. The work that Cécile Hardebolle and CEDE are undertaking perfectly reinforces these notions, and has initiated the reflection upstream of my lesson“ says Flammarion.

Embedded in a larger effort to integrate ethics education in engineering, this learning tool is part of swissuniversities‘ „P-8 Digital Skills“ program. It is available to all EPFL teachers, adding to the series of interventions developed by CEDE and the Teaching Support Center (CAPE), both part of the Center LEARN for Learning Sciences.

A first for Switzerland

Endowed with 77’000 euros, this is the most important prize for research in subject didactics and honors outstanding scientists from the regions of Germany, Austria and Switzerland. Under the patronage of the Federal Minister of Education and Research, Bettina Stark-Watzinger, it recognizes contributions to the research and development of innovative teaching concepts, with this year, a focus on the use of digital tools as a resource for STEM education (science, technology, engineering, and mathematics).

Professor Mondada distinguished himself with the educational robot Thymio II and its application to the teaching of computational thinking, becoming the first Swiss scientist to receive the prestigious award.

Thymio II was chosen as one of eight shortlisted concepts after a rigorous selection process by a panel of Germany’s top didactics experts, and was announced as the winning project during the award ceremony held Friday in Frankfurt, which was attended by several hundred prominent German officials.

In accordance with Polytechnische Gesellschaft tradition, the prize is awarded for projects that hold strong transfer potential. The program’s goal is to foster young people’s curiosity and spirit of experimentation and in the long run, the winning projects will contribute to the city’s educational landscape.

Collaboration encoded in its DNA

While there are currently 80,000 Thymios on the market, with 70% of them being used in schools outside of our borders, this initiative did not come about overnight. In fact, it is primarily thanks to the invaluable partnerships and close collaboration with the cantons that it was able to gain popularity among Swiss children.

„Thymio is a project that was progressively built from the ground up, it was a journey enabled by a range of committed partners,“ Professor Mondada emphasizes. „These results are the outcome of our collaboration with the cantons, who were instrumental in providing a field for us to develop, test, co-construct, and deploy our concepts together with teachers.We focused on computational thinking concepts in particular and were able to scientifically assess their reactions as well as those of the students.“

Thymio, which recently celebrated its eleventh birthday, was designed from the get go with collaboration encoded in its DNA. EPFL, ECAL, ETHZ, INRIA, the NCCR Robotics, and cantonal authorities have all gathered around the project over the years. The canton of Geneva was the first to introduce the Thymio robot into schools, followed by Valais, which also made this tool available to teachers who showed interest in it. With the EduNum project, the canton of Vaud included it in its project to introduce digital education in every school, enabling the study of teachers‘ reactions on a large scale. In its current implementation of digital education, the canton of Neuchâtel makes use of the link between Thymio and computational thinking. Furthermore, Thymio has been included in the canton of Bern’s „MINT mobil“ approach, which aims to reach all schools in the canton. Finally, Ticino, in collaboration with the SUPSI and the USI, has made significant contributions to the implementation of computational thinking models and the introduction of Thymio to Ticino teachers.

This diffusion was facilitated by Thymio’s highly innovative design. Designed from the outset to be open source, the educational robot is equipped with sensors and actuators and allows anyone to learn basic programming skills through a tangible framework allowing users to grasp and manipulate the robot directly. A very intuitive visualization of the robot’s functionalities allows for a quick understanding of its mechanisms. Because of its open and easily repairable design, it is a tool that adheres to several sustainability principles. Today, it is at the center of many communities, each one more innovative and creative than the other, that all feed its ecosystem with a plethora of educational resources. This ecosystem is supported by the non-profit association Mobsya, which produces the robot and maintains the software tools around Thymio.

Thirty years of translational research

The development of these technical and educational elements has been guided by continuous translational research in robotics and learning sciences, a true strength of the EPFL’s Center LEARN. Thymio has been the subject of numerous studies on its impact in a variety of contexts, the results of which have led to over twenty international scientific publications and significant visibility in these two areas of research.

Francesco Mondada, a true pioneer in the field, has been designing tabletop robots for education and research as early as the early 1990s and has infused Thymio with nearly three decades of research findings, making it the ideal tool for teaching the fundamentals of computer science.

„Using it in a problem-solving setting fosters a link between the teaching of computer science and the teaching of traditional disciplines and cross-curricular skills such as thinking, creativity, or collaboration. This linkage facilitates teacher adoption as well as the introduction of elements of computer science at the school level,“ says the Professor.

Following the announcement of the ranking by the jury, he remains unwaveringly humble: „It’s quite incredible considering that Germany is not a country where Thymio has spread much (it is more present in French-speaking countries), it is an international prize that seeks to support innovation in didactics specifically in STEM fields. We’ve received other awards, but never on this scale; this is truly an honor and surely an unforgettable moment“ he adds.

Like many of the teaching staff at the EPFL, Jean-Cedéric Chapelier had a number of long lecture recordings. These were recorded in the CE amphitheaters on campus and lasted for the duration of the lecture, around 1hr45.

At CEDE we were able to easily split up these lengthy recordings into a number of shorter more digestible videos for his students. All Jean-Cedric had to do was provide us with some suggested times for splitting the recordings.

The result of this collaboration was a list of more than 100 concise videos that switched between a clear view of the slides and Jean-Cedric presenting to the class. These videos can be used in a range of different ways for years to come. Revision resources, teaching videos or even the base for creating an online course.

So if you have a long list of lecture recordings, extract all of their untapped potential by getting in contact with the video team at CEDE.

For engineers, little can be more satisfying than turning your ideas into working objects you can see and touch. That’s exactly what two groups from EPFL’s chemistry and chemical engineering section recently got to experience. After six months of hard work on the DLL Molecular – Chemical Engineering program, the dozen or so Master’s students got to unveil their creations: a self-heating food box and a self-cooling vaccine container.

It all started with the Chemical Engineering Product Design class in the fall semester. Interested students were given the option of signing up for the Chemical Engineering Lab & Project in the spring semester, giving them a chance to build a prototype of their initial blueprints. “The class has a dual purpose: working through the design process from start to finish, and rethinking everyday products to make them more sustainable,” says Prof. Jeremy Luterbacher, who heads EPFL’s Laboratory of Sustainable and Catalytic Processing (LPDC). The approach is about giving students the freedom to unleash their creativity, forge a sense of belonging to the School and their group, and develop cutting-edge engineering skills.

Students who signed up for the class chose from a predefined list of projects and started by developing a technical blueprint. The participants who showed the highest levels of motivation and interest were then selected to build a prototype. While EPFL has offered the desk-based class for many years, this was the first time students were given an opportunity to prototype their designs. The process inevitably threw up unexpected challenges but proved to be a valuable learning experience – one that produced compelling findings.

“The groups not only built something that works, but also performed a vast array of calculations and methods – it was an elegant exercise in chemical engineering,” explains Luterbacher, who was impressed with how the students applied themselves. Their hard work is embodied in the prototypes they built, each of which has a special feature. The self-heating food box, made from 3D-printed resin, is stylishly designed and decorated, while the self-cooling vaccine container – dubbed Frigivax – has its own logo, carefully engraved on the lid.

A hotbed of ideas

The students were challenged to come up with designs that used neither microwaves nor electricity. The group behind the self-heating food box opted for a block filled with calcium oxyde – better known as slaked lime. It’s used widely in the construction industry and reacts with water to form calcium hydroxyde and release heat. The students dropped the block into the box, added a generous dose of water, placed the food on top, and then closed the lid. After 10 minutes, the food was heated through and ready to eat, with tests showing that the temperature inside the box could reach as much as 100°C. What’s more, the block’s contents can be recycled and reused – a fact that further bolsters the project’s environmental credentials.

Getting to this point wasn’t easy, however. For the four students in the group, developing the food-heating system meant exercising their gray matter. “We spent at least two months doing nothing but brainstorming,” says Jana Lukic. According to fellow student Lorenzo Mazzoli, the group “put around a hundred ideas on the table.” Group member Ting-Wei Weng adds: “We considered every possible avenue, narrowing the field down to five or six realistic options.” Maxime Brunisholz says the students “even explored crazy-sounding options like whacking the food to generate heat.”

One challenge the students faced stemmed from the properties of calcium hydroxide, which can cause chemical burns when it comes into contact with skin. They ultimately bypassed this problem by adding citric acid.

The oral exam was a unique experience: I ended up tasting the end product

Hands-on building

The group behind the self-cooling vaccine container faced a dual challenge. In order to maintain a temperature of between 2°C and 10°C for several days on end, they had to come up with a way to both absorb heat and keep the box insulated.

For the first challenge, the students drew on a property that’s common to most materials. Compounds in a solid state generate cold when they melt; those in a liquid state generate heat when they freeze. The group combed through research by NASA into the physical properties of cooling materials, looking for those that operated within their target temperature range. In the end, they opted for tetradecane, which they fashioned into ice packs to sit at the bottom of the container.

Tetradecane is a hydrocarbon that melts at around 6°C, keeping the temperature inside the box constant at around the same level. “If it’s 40°C outside, the system will last twice as long as if it’s 25°C, even with the same amount of tetradecane,” says Raphaël Finizola. “We can customize the containers by adding more or less of the material as required.”

To keep the box insulated, the group initially wrapped it in insulating foam. But that wasn’t enough, so they also added vacuum insulation panels. “Combining tetradecane with the insulating materials produced the result we were looking for: the temperature inside the container rose from 2°C to 6°C in just four days,” says Lise Boitard-Crépeau. “Self-cooling systems like these already existed, but ours is the first one that works for so long.” Although the Frigivax container is designed to carry vaccines, it could also be used to transport food, or potentially organs. What’s more, all the components can be recycled and reused.

“It was an interesting experience,” says Simon Baillet. “As self-reliant engineers, we had to do everything ourselves: design the system and choose the materials – while keeping within our assigned budget. We really learned a lot in the process. And it was good to do some hands-on building in what’s otherwise a very science-focused program.”

The three new Master’s programs are offered at the intersection between disciplines. This allows students with varied backgrounds in science and engineering to have the opportunity to acquire a comprehensive set of skills to work in the MedTech, pharma and health care sectors (Master’s in Neuro-X) or become the main actors of the “quantum revolution” (Master’s in Quantum Science and Engineering). The Master’s in Statistics aims to provide scientists with the expertise and crucial skills for sound reasoning in the data-rich world, making them sought-after statisticians and data analysts.

EPFL constantly adapts its education offer to the new developments in science and engineering, as well as to the evolution of our society, the emerging needs of its economy as well as its numerous challenges

The Master’s in Neuro-X

Engineers in Neuro-X build their expertise on science, technology and computation. Their multidisciplinary expertise complements the fundamental skills of engineers and medical-domain specialists by a strong technological component, making them not only highly demanded and valued professionals in neurotechnology, but also preparing them for research in neuroscience-related fields. The study program includes several projects in labs, thereby providing students with a practical dimension and real research experience.

Prof. Dimitri Van De Ville, the Master’s program director, foresees that graduates will have an interdisciplinary profile enabling them to see the big picture in terms of complex systems, combined with a realistic perspective on what it means to develop a product or engage into research. This will make them key actors able to interact with experts at the intersection of domains.

The Master’s in Quantum Science and Engineering

Quantum science and technology is bringing a paradigm shift in the way we treat, communicate, measure and compute data, affirms the Master’s program director Prof. Nicolas Macris. He adds that to address this new shift, EPFL aims to form quantum science engineers who, thanks to their multidisciplinary profile, are able to thrive at the forefront of this “new technological revolution” and to pursue a career in quantum science or in the information technology sector as well as in the industry at large.

The Master’s in Statistics

In a world that is increasingly data-driven, industry relies on statisticians and data analysts who are able to navigate the data flood. Statistical expertise is now essential in nearly all domains: economics, finance, government, science, health, and social sciences, and the list goes on. “With the Master’s in Statistics, EPFL aims to train students with a scientific/engineering background in cutting-edge statistical methodology, in order to develop a mastery of statistical thinking, visualization and computation, and data analysis” – so says Prof. Joachim Krieger, the director in charge of the program. Teamwork and communication skills are also important aspects that the program strengthens, in order to enable graduates to integrate and apply their skills in the manifold fields of application of statistics.

For more information, see https://www.epfl.ch/education/master/programs/

Thanks to Jupyter Notebooks, students can solve structural engineering problems by watching structural deformation as it happens, understand signal processing with the help of sounds, music or images, and grasp abstract concepts in physics – all in a simple, accessible manner. The notebooks’ digital environment combines computing power with course content so that students can practice computational thinking. This bolsters their conceptual reasoning and expands their programming skillset. Teachers use them to run virtual demonstrations during class and for assignments that students can work on remotely. In addition, the notebooks’ interactive interface enables students to work out problems and deepen their knowledge.

The project to develop the use of Jupyter notebooks in education at EPFL got underway in 2019. “We’d already been thinking about it for some time,” says Patrick Jermann, Executive Director of EPFL’s Center for Digital Education. “We discussed it with Pierre Vandergheynst, who was EPFL’s Vice President for Education at the time, since incorporating computational thinking into our degree programs was in line with EPFL’s strategic educational objectives. And Jupyter notebooks make it possible to use computational methods to help students understand concepts from a variety of disciplines.”

The Jupyter notebooks are an open-source technology born in the US. “They were originally named IPython Notebooks, after the first programming language they supported,” says Cécile Hardebolle, pedagogical advisor in charge of the project at EPFL. “Then came the Jupyter project, whose name is a contraction of Julia, Python and R – the first three languages implemented in the platform. Today, there are many more.”

To support the use of Jupyter notebooks at EPFL, the project team first had to set up the required IT infrastructure and adapt it to users’ requirements. This task fell to Pierre-Olivier Vallès, a systems engineer at EPFL. “Assembling the various components and getting them to work together was a massive undertaking,” he says. “Our goal was to create a system that could meet EPFL’s needs and fit in with our other IT systems, like the Moodle learning platform and our MOOCs services.”

The Jupyter Notebook for education service was rolled out gradually with help from teachers who were interested in the new teaching method and from those who had already been using the Notebooks for research purposes. Cécile Hardebolle explains: “The real technical challenge was to adapt the system to specific teaching requirements. For example, if a chemistry professor wanted to demonstrate computational chemistry and needed a given library, Pierre-Olivier would add it. We’re always on the lookout for new libraries and extensions that could fit well in an educational setting.” A range of fields taught at EPFL – such as chemistry, machine learning and geographic information systems – can benefit from the notebooks’ digital environment.

Although using Jupyter Notebooks is easy, installing the servers to support them isn’t. The added value provided by EPFL is that, thanks to noto, the JupyterLab centralized platform for education, the Notebooks can be used without downloading and installing special software. This saves teachers a considerable amount of time and means that students can work from anywhere, even if they don’t have a powerful laptop.

The effort to roll out the notebooks service at EPFL has paid off: since 2019, over 5,500 individuals have connected to noto, including professors and users from other universities who are curious about the technology. There are some 2,600 regular users, meaning the system needs to be robust enough to handle a number of simultaneous queries. “If a class of 30 students logs in at once, that must work fine,” says Cécile Hardebolle. “And if 50, 100 or 200 students try to connect at 8:15 am, all the servers must be up and running within 5 minutes.”

Prof. Martin describes his teaching method as connecting the chalkboard with the lab bench. Flattered to have won this year’s best teacher award for the microengineering section, he nevertheless insists that “I feel rewarded every day I teach at EPFL.”

And he’s been reaping those rewards for 17 years now. However, he admits it hasn’t always been easy. “I lacked experience at first,” he says. “And since I tend to be shy, people said I spoke too softly.” But today – thanks to the help of a voice coach, encouraging feedback from students, and a few semesters of experience – Prof. Martin is delighted to get back in front of the classroom each fall. “Every new school year feels like the first time, since each class is so different,” he says.

Capturing students’ attention

He adapts to that difference by tailoring his lectures to the class’s level of knowledge every year. Prof. Martin has come a long way from his days as a child, when he stuck his finger in an electrical socket to see how it worked; today he teaches electricity to first-year Bachelor’s students. “I give my theory class entirely on a chalkboard. It’s a good way to capture students’ attention,” he says. He also passes out thick stacks of handouts, printed on a single side to encourage students to take notes and write out their solutions to problems. “That helps them put in the effort needed to understand the material,” he explains. As a further incentive for students to take notes, they can bring their handouts with them to exams.

Hands-on experiments are another method Prof. Martin uses to teach his first-year students. The experiments are designed to closely parallel the theory studied in class and to introduce or illustrate specific concepts. “By bringing a practical dimension to arduous subjects like physics and mathematics, I want the students to experience lightbulb moments,” he says. The combination of theory and first-hand experience gives students a better – and longer-lasting – grasp of concepts.

Thinking ahead

“The first year of a Bachelor’s program can be difficult and highly abstract,” says Prof. Martin. “But it’s important for students to understand that the fundamental topics they’re studying now are necessary to make robots fly.” To help students make that connection, he has them run experiments in EPFL’s Discovery Learning Labs (DLLs) – or as he likes to call them, “showcase labs.” His electrical engineering students can work with Prof. Martin as well as other teachers to explore the wide range of concrete applications for things like signal processing, embedded systems, photonics, acoustics and power.

Prof. Martin also teaches an optical engineering class for third-year Bachelor’s students along with one Master’s-level class. Here too, he wants students to get as much hands-on experience as possible, either by conducting experiments in DLLs or by running computer simulations on MatLab in order to model different types of optical systems. The student evaluations he receives are generally excellent, with 97–100% of them containing positive feedback along with praising remarks.

Prof. Martin led the microengineering section from 2016 to 2020. His is a relatively young field born from the merger of electrical and mechanical engineering around 20 years ago. “The study plan tends to change chaotically based on the opportunities at hand,” he says. He took advantage of his leadership role to revamp the Bachelor’s and Master’s programs – in close cooperation with fellow professors – by scaling back the number of first-year classes and introducing a common thread in subsequent semesters.

A MOOC in the works

Did the pandemic change Prof. Martin’s approach? It threw up several challenges, of course, which he navigated by recording some 180 videos – including of chalkboard lessons – which have been viewed on Switchtube over 18,500 times. The pandemic also made one thing clear. “I’m fully convinced that teaching in person is the best way to go,” he says. “Although I must admit that there were some benefits to structuring the content of a class to make it better suited for online lessons. I now plan to develop a MOOC version of my optical engineering class.”