This week, Dr. Lanusse will review different established practices for Bayesian Neural Networks (e.g. Bayes by Backprop, MC Dropout) and their usefulness to model epistemic uncertainties in network predictions. With these notions in place, we will take a look at how these tools have been used in astronomical data analysis, for time series classification (https://arxiv.org/abs/1901.06384), and physical parameter inference from images (https://arxiv.org/abs/1708.08843). Full details about this meeting will be posted here: https://bids.github.io/MLStatsForum/.
The Berkeley Statistics and Machine Learning Forum meets biweekly to discuss current applications across a wide variety of research domains and software methodologies. Hosted by UC Berkeley Physics Professor and BIDS Senior Fellow Uros Seljak, these active sessions bring together domain scientists, statisticians and computer scientists who are either developing state-of-the-art methods or are interested in applying these methods in their research. Practical questions about the meetings can be directed to BIDS Fellow Francois Lanusse. All interested members of the UC Berkeley and LBL communities are welcome and encouraged to attend. To receive email notifications about the meetings and upvote papers for discussion, please register here.
I am a Data Science Fellow at BIDS and the Berkeley Center for Cosmological Physics exploring the intersection between Cosmology, Statistics, and Machine Learning. My research has been focused on measuring and exploiting the gravitational lensing effect (in which distant galaxies appear distorted due to the presence of massive structures along the line of sight) with the development of novel tools and methodologies based on sparse signal representations, convex optimization, and deep learning.
I am an active member of the Large Synoptic Survey Telescope (LSST) Dark Energy Science Collaboration which aims at answering pressing questions about the nature of Dark Energy, a force thought to drive the accelerated expansion of the Universe. LSST will observe billions of galaxies over a period of ten years, measuring in particular the lensing effect in great details to constrain cosmological models. The unprecedented scale and complexity of these modern cosmological surveys involve a number of outstanding challenges which drive most of my research into new methodologies impacting different stages of the science analysis, from image processing to the statistical inference of cosmological parameters.
Before joining Berkeley I have been working as a postdoctoral researcher within the McWilliams Center for Cosmology at Carnegie Mellon University, after completing a PhD in astrophysics at CEA Saclay near Paris in 2015.