Machine Learning

An international team of scientists led by EPFL has developed a system that combines information from the brain’s connectome – the “wiring” between neurons – and machine learning to assess and predict the outcome of stroke victims.

Chemical engineers at EPFL have developed a machine-learning model that can predict a compound’s oxidation state, a property that is so essential that many chemists argue it must be included in the periodic table.

Artificial intelligence and machine learning are key technologies for science, business, and society – but how transparent are their decisions? Andreas Krause, the 2021 recipient of the Rössler Prize and Chair of the ETH AI Center, shares his thoughts on the opportunities and challenges with trustworthy artificial intelligence.

From robots that offer solace to algorithms that help judges make fact-based decisions, robotics and machine learning are entering new domains that were once the preserve of humans.

EPFL scientists in a European collaboration propose a personalized protocol for optimizing stimulation of optic nerve fibers, for the blind, which takes into account feedback from the viewer’s brain. The protocol has been tested on artificial neural networks known to simulate the physiology of the entire visual system, from the eye to the visual cortex. The stimulation protocol will be tested in clinical trials with partners in Rome.

How do you teach machines to learn for themselves and support people? Andreas Krause is one of Europe’s leading machine learning researchers. His approaches combine mathematical elegance with a sense of social responsibility. He has now been awarded the Rössler Prize, ETH Zurich’s most generous research award.