Research initiatives

The EPFL AI Center is the hub for large-scale AI research promoting collaboration across disciplines in AI research and its real-world applications at EPFL.

With more than 95 professors in AI and machine learning, EPFL is at the forefront of advancing modern AI across various sectors. 

Our strategic research initiatives span a broad spectrum of advanced AI domains, addressing both fundamental and applied challenges. The areas listed below highlight our main focus, though other topics are also explored. 

AI for Health

The AI for Health strategic research initiative is dedicated to developing AI to advance healthcare and biomedical sciences. This involves building partnerships and long term research projects with the local and global healthcare ecosystem to foster collaboration and information sharing between health experts and AI researchers with the goal of improving prevention, patient outcomes and care. Projects could include advancing AI research and tools for clinical use, drug discovery, digital health monitoring, global health, and personalised medicine.

AI for Science

The AI for Science strategic research initiative focuses on projects that harness AI to address fundamental scientific problems, unlocking new venues of interdisciplinary research with the potential to benefit society. By leveraging AI across diverse fields such as biology, chemistry, physics, materials science, and environmental science, this initiative accelerates discovery and innovation. Examples of projects include the development of AI models to predict the properties of new materials, understand complex biological systems, or facilitate a deeper understanding of intricate phenomena such as climate dynamics and molecular interactions.

AI for Robotics

The AI for Robotics strategic research initiative aims to enhance the capabilities of robots through the integration of novel AI techniques, such as using reinforcement and imitation learning, enabling robots to learn from their environment, adapt to new tasks, or interact efficiently with humans. These advancements can be applied across a wide range of domains, from predictive maintenance and robotic manipulation to human-robot collaboration and bio-inspired robotics.

AI for Education

The AI for Education strategic initiative focuses on using AI to enhance learning experiences and educational outcomes. This includes developing intelligent tutoring systems, personalised learning platforms, or devising new approaches to student evaluation, with the aim of making education more personalised, accessible, and effective for learners of all ages and backgrounds.

AI for Sustainability

The AI and Sustainability strategic research initiative focuses on projects that develop AI to address critical environmental challenges and promote sustainable development, as well as projects aimed at making AI itself more sustainable and efficient. Topics of interest include developing AI models to optimise energy consumption, minimise waste, enhance the efficiency of renewable energy sources, and improve the overall efficiency of AI systems.

AI Fundamentals

The AI Fundamentals strategic research initiative focuses on advancing the core theoretical foundations of AI while deepening our understanding of AI models through rigorous empirical experimentation. This includes exploring and designing new algorithms, enhancing learning models, and investigating the mathematical underpinnings underlying AI systems. This research will lay the groundwork for future AI applications across various domains, in collaboration with other research initiatives such as AI for Health and AI for Science.

AI for Society

The AI for Society strategic research initiative aims at ensuring that AI technologies are developed and deployed in ways that are ethical, safe, trustworthy and aligned with societal values, while balancing transparency and privacy concerns. This initiative is transversal in nature, with its principles being integrated into each of the other strategic research initiatives in a context relevant manner.

Share