Deep Learning
A new tool for protein sequence generation and design
Research
EPFL
EPFL researchers have developed a new technique that uses a protein language model for generating protein sequences with comparable properties to natural sequences. The method outperforms traditional models and offers promising potential for protein design.
Time to get social: tracking animals with deep learning
TechTransfer
EPFL
Researchers at EPFL have made strides in computer-aided animal tracking by expanding their software, DeepLabCut, to offer high-performance tracking of multiple animals in videos.
Discover the underwater world
Research
Eawag
Our lakes, rivers and streams are teeming with the smallest creatures, plants and bacteria that are barely visible to the naked eye, if at all. An underwater camera makes it possible to observe and identify the species of these creatures in real time.
Deep-learning-based image analysis is now just a click away
Research
EPFL
Under an initiative by EPFL’s Center for Imaging, a team of engineers from EPFL and Universidad Carlos III de Madrid have developed a plugin that makes it easier to incorporate artificial intelligence into image analysis for life-science research. The plugin, called deepImageJ, is described in a paper appearing today in Nature Methods.
EPFL algorithm in the world's most popular deep learning software
Research
EPFL
PyTorch is used for countless AI applications ranging from Tesla’s Autopilot to Facebook’s translation software and its latest version, just launched, features an EPFL communication-efficient training algorithm that even helps the planet. 
Page 2 of 2