לוח שנה משולב

The Atlantic Meridional Overturning Circulation (AMOC) is a circulation pattern of great climatic importance. Its northward heat flux at the upper water column moderates European winter climate, and its descending branch captures atmospheric CO2 into the deep ocean, hence buffering the anthropogenically induced rise in global temperature.
The Deep Western Boundary Current (DWBC) has classically been considered to be the main AMOC conduit southward at depth. However, tracer data have shown in recent decades that the DWBC "leaks" most of its material to the ocean interior in a small region of the North Atlantic, and that this leaked material continues southward in different, complex routes. These pathways and their causes are still little-explored and not well understood.
In this talk I will present analysis of the DWBC leakiness properties and dynamics, based on existing datasets of passively drifting floats, a new high resolution regional numerical model, and theoretical analysis. Several alternative mechanisms of leakiness are considered, and a novel finding is that a leading cause for the leakiness is inertial separation of the current from the seafloor, near underwater capes. The role of eddies and their interaction with the separation process is investigated as well. Implications for the robustness of the deep AMOC pathways are discussed.

The presentation will discuss recent evidence supporting a role for the gut-brain axis in controlling brain circuits involved in reward. It will be argued that sensory neurons of vagus nerve function as reward neurons. Via defined brainstem targets, vagal signals dopaminergic brain reward circuits in midbrain. The mapping of these circuits opens a window into how signals generated by internal body organs give rise to motivated and emotional behaviors.

Zoom link to join-
https://weizmann.zoom.us/j/96608033618?pwd=SEdJUkR2ZzRBZ3laUUdGbWR1VFJTdz09

Meeting ID: 966 0803 3618
Password: 564068

Host: Dr. Rita Schmidt rita.schmidt@weizmann.ac.il tel: 9070

Zoom Lecture: https://weizmann.zoom.us/j/98819686427?pwd=algvMEJUNHdvaFppNS9xVzlTUkhYQT09
Passcode: 551107

Many functions of RNA depend on rearrangements in secondary structure that are triggered by external factors, such as protein or small molecule binding. These transitions can feature on one hand localized structural changes in base-pairs or can be presented by a change in chemical identity of e.g. a nucleo-base tautomer. We use and develop R1ρ-relaxation-dispersion NMR methods for characterizing transient structures of RNA that exist in low abundance (populations

mRNA based vaccines prevent COVID-19 infections, putting them at the forefront of the current global fight against the COVID-19 pandemic. The scientific and clinical development of mRNA medicines, which began in ernest only ~10 years ago, has the potential to not only change the landscape of infectious disease vaccines but to also impact the treatment of cancer, genetic metabolic, autoimmune, and cardiovascular diseases. This talk will review the translational medicine approach to the research and development of both infectious disease vaccines, as exemplified by COVID-19 vaccine Moderna, as well as other applications of mRNA medicines currently in clinical development.
ᐧ

The Burger-Sarnak method shows that the restriction of an automorphic representation of a reductive group to a reductive subgroup has automorphic support. Clozel has conjectured a qualitative refinement of this result, which was first verified and quantified in the GLn case by Venkatesh. In this talk I will describe my thesis which extended this result to classical groups.

Deep Learning has always been divided into two phases: Training and Inference. The common practice for Deep Learning is training big networks on huge datasets. While very successful, such networks are only applicable to the type of data they were trained for and require huge amounts of annotated data, which in many cases are not available. In my thesis (guided by Prof. Irani), I invented ``Deep Internal Learning''. Instead of learning to generally solve a task for all inputs, we perform ``ad hoc'' learning for specific input. We train an image-specific network, we do it at test-time and on the test-input only, in an unsupervised manner (no label or ground-truth). In this regime, training is actually a part of the inference, no additional data or prior training is taking place. I will demonstrate how we applied this framework for various challenges: Super-Resolution, Segmentation, Dehazing, Transparency-Separation, Watermark removal. I will also show how this approach can be incorporated to Generative Adversarial Networks by training a GAN on a single image. If time permits I will also cover some partially related works.

Links to papers:
http://www.wisdom.weizmann.ac.il/~vision/zssr
http://www.wisdom.weizmann.ac.il/~vision/DoubleDIP
http://www.wisdom.weizmann.ac.il/~vision/ingan
http://www.wisdom.weizmann.ac.il/~vision/kernelgan
https://semantic-pyramid.github.io/
https://arxiv.org/abs/2006.11120
https://arxiv.org/abs/2103.15545