CHRISTOPH JOHANN FEINAUER
Courses a.y. 2022/2023
30539 COMPUTER SCIENCE - MODULE 1 (INTRODUCTION TO COMPUTER SCIENCE AND PROGRAMMING)
30586 MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE LAB
I teach a master course in advanced machine learning, one bachelor course in AI and one on programming using C.
I am an Assistant Professor of Computer Science at Bocconi University in Milan, Italy. I obtained my PhD in the context of the Marie-Curie program Netadis from the Polytechnic University of Turin, with research stays at the UPMC Paris and the KTH Stockholm. My work focuses on generative models for protein sequence data and explainable deep learning. After doing a postdoc in Martin Weigt’s group in Paris, I worked as a Machine Learning Scientist in a deep learning startup in Berlin and then joined Bocconi University in 2019.
My research interests are mainly generative models for protein sequence models, protein design and explainable deep learning with applications for deep models trained on biological data. I am especially interested in the prediction of pathogencity of genomic mutations using deep learning and the interpretation of the results in terms of the underlying biology. I also work on the structure of loss landscapes of neural networks, using methods derived from statistical physics.
Reconstruction of Pairwise interactions using energy-based models
JOURNAL OF STATISTICAL MECHANICS: THEORY AND EXPERIMENT, 2021
Interpretable Pairwise Distillations for Generative Protein Sequence Models
Workshop on Machine Learning in Structural Biology
Entropic gradient descent algorithms and wide flat minima
Proceedings of International Conference on Learning Representations, 2021
Reconstruction of pairwise interactions using Energy-Based Models
Proceedings of Machine Learning Research vol 145: 2nd Annual Conference on Mathematical and Scientific Machine Learning, 9999
Proceedings of Machine Learning Research.
Natural representation of composite data with replicated autoencoders.
Architecture of a mammalian glomerular domain revealed by novel volume electroporation using nanoengineered microelectrodes
NATURE COMMUNICATIONS, 2018
Inverse statistical physics of protein sequences: a key issues review
REPORTS ON PROGRESS IN PHYSICS, 2018
Context-aware prediction of pathogenicity of missense mutations involved in human disease
Mutator genomes decay, despite sustained fitness gains, in a long-term experiment with bacteria
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2017
Inter-protein sequence co-evolution predicts known physical interactions in bacterial ribosomes and the trp operon
PLOS ONE, 2016
Improving contact prediction along three dimensions
PLOS COMPUTATIONAL BIOLOGY, 2014
Fast and accurate multivariate Gaussian modeling of protein families: predicting residue contacts and protein-interaction partners
PLOS ONE, 2014
Zinc finger proteins and the 3D organization of chromosomes
Organisation of chromosomes, 2013