University of Tübingen

About: University of Tübingen is a education organization based out in Tübingen, Germany. It is known for research contribution in the topics: Population & Immune system. The organization has 40555 authors who have published 84108 publications receiving 3015320 citations. The organization is also known as: Eberhard Karls University & Eberhard-Karls-Universität Tübingen.

On Adaptive Attacks to Adversarial Example Defenses

Abstract: Adaptive attacks have (rightfully) become the de facto standard for evaluating defenses to adversarial examples. We find, however, that typical adaptive evaluations are incomplete. We demonstrate that thirteen defenses recently published at ICLR, ICML and NeurIPS---and chosen for illustrative and pedagogical purposes---can be circumvented despite attempting to perform evaluations using adaptive attacks. While prior evaluation papers focused mainly on the end result---showing that a defense was ineffective---this paper focuses on laying out the methodology and the approach necessary to perform an adaptive attack. We hope that these analyses will serve as guidance on how to properly perform adaptive attacks against defenses to adversarial examples, and thus will allow the community to make further progress in building more robust models.

A White Paper on keV sterile neutrino Dark Matter

Abstract: We present a comprehensive review of keV-scale sterile neutrino Dark Matter, collecting views and insights from all disciplines involved—cosmology, astrophysics, nuclear, and particle physics—in each case viewed from both theoretical and experimental/observational perspectives. After reviewing the role of active neutrinos in particle physics, astrophysics, and cosmology, we focus on sterile neutrinos in the context of the Dark Matter puzzle. Here, we first review the physics motivation for sterile neutrino Dark Matter, based on challenges and tensions in purely cold Dark Matter scenarios. We then round out the discussion by critically summarizing all known constraints on sterile neutrino Dark Matter arising from astrophysical observations, laboratory experiments, and theoretical considerations. In this context, we provide a balanced discourse on the possibly positive signal from X-ray observations. Another focus of the paper concerns the construction of particle physics models, aiming to explain how sterile neutrinos of keV-scale masses could arise in concrete settings beyond the Standard Model of elementary particle physics. The paper ends with an extensive review of current and future astrophysical and laboratory searches, highlighting new ideas and their experimental challenges, as well as future perspectives for the discovery of sterile neutrinos.

The senescence-associated secretory phenotype induces cellular plasticity and tissue regeneration

Abstract: Senescence is a form of cell cycle arrest induced by stress such as DNA damage and oncogenes. However, while arrested, senescent cells secrete a variety of proteins collectively known as the senescence-associated secretory phenotype (SASP), which can reinforce the arrest and induce senescence in a paracrine manner. However, the SASP has also been shown to favor embryonic development, wound healing, and even tumor growth, suggesting more complex physiological roles than currently understood. Here we uncover timely new functions of the SASP in promoting a proregenerative response through the induction of cell plasticity and stemness. We show that primary mouse keratinocytes transiently exposed to the SASP exhibit increased expression of stem cell markers and regenerative capacity in vivo. However, prolonged exposure to the SASP causes a subsequent cell-intrinsic senescence arrest to counter the continued regenerative stimuli. Finally, by inducing senescence in single cells in vivo in the liver, we demonstrate that this activates tissue-specific expression of stem cell markers. Together, this work uncovers a primary and beneficial role for the SASP in promoting cell plasticity and tissue regeneration and introduces the concept that transient therapeutic delivery of senescent cells could be harnessed to drive tissue regeneration.

D-NeRF: Neural Radiance Fields for Dynamic Scenes

Abstract: Neural rendering techniques combining machine learning with geometric reasoning have arisen as one of the most promising approaches for synthesizing novel views of a scene from a sparse set of images. Among these, stands out the Neural radiance fields (NeRF), which trains a deep network to map 5D input coordinates (representing spatial location and viewing direction) into a volume density and view-dependent emitted radiance. However, despite achieving an unprecedented level of photorealism on the generated images, NeRF is only applicable to static scenes, where the same spatial location can be queried from different images. In this paper we introduce D-NeRF, a method that extends neural radiance fields to a dynamic domain, allowing to reconstruct and render novel images of objects under rigid and non-rigid motions from a \emph{single} camera moving around the scene. For this purpose we consider time as an additional input to the system, and split the learning process in two main stages: one that encodes the scene into a canonical space and another that maps this canonical representation into the deformed scene at a particular time. Both mappings are simultaneously learned using fully-connected networks. Once the networks are trained, D-NeRF can render novel images, controlling both the camera view and the time variable, and thus, the object movement. We demonstrate the effectiveness of our approach on scenes with objects under rigid, articulated and non-rigid motions. Code, model weights and the dynamic scenes dataset will be released.