Conferences - Seminars

  12:00 BSP 234

Deep learning has been immensely successful at a variety of tasks, ranging from classification to artificial intelligence. Yet why it works is unclear. Learning corresponds to fitting training data, which is implemented by descending a very high-dimensional loss function. Two central questions are (i) since the loss is a priori not convex, why doesn't this ... Read more about "On the geometry of the landscape underlying deep learning"
  14:00-15:00 BSP 407

Microscopy has fueled biological discoveries for centuries, but technical progress during the last decades has greatly expanded the type and quantity of biological information that can be revealed by imaging. Advances in instrumentation, labeling and computation are driving an imaging data tsunami on par with that of DNA sequencing. We illustrate this trend and discuss ... Read more about "Computational imaging and modeling for cellular and nuclear biology"
By Christophe Zimmer