University of Fribourg

About: University of Fribourg is a education organization based out in Fribourg, Freiburg, Switzerland. It is known for research contribution in the topics: Population & Glacier. The organization has 6040 authors who have published 14975 publications receiving 542500 citations. The organization is also known as: UNIFR & Universität Freiburg.

Morpheus: towards automated SLOs for enterprise clusters

Abstract: Modern resource management frameworks for large-scale analytics leave unresolved the problematic tension between high cluster utilization and job's performance predictability--respectively coveted by operators and users. We address this in Morpheus, a new system that: 1) codifies implicit user expectations as explicit Service Level Objectives (SLOs), inferred from historical data, 2) enforces SLOs using novel scheduling techniques that isolate jobs from sharing-induced performance variability, and 3) mitigates inherent performance variance (e.g., due to failures) by means of dynamic reprovisioning of jobs. We validate these ideas against production traces from a 50k node cluster, and show that Morpheus can lower the number of deadline violations by 5× to 13×, while retaining cluster-utilization, and lowering cluster footprint by 14% to 28%. We demonstrate the scalability and practicality of our implementation by deploying Morpheus on a 2700-node cluster and running it against production-derived workloads.

Learning of sequential movements by neural network model with dopamine-like reinforcement signal.

Abstract: Dopamine neurons appear to code an error in the prediction of reward. They are activated by unpredicted rewards, are not influenced by predicted rewards, and are depressed when a predicted reward is omitted. After conditioning, they respond to reward-predicting stimuli in a similar manner. With these characteristics, the dopamine response strongly resembles the predictive reinforcement teaching signal of neural network models implementing the temporal difference learning algorithm. This study explored a neural network model that used a reward-prediction error signal strongly resembling dopamine responses for learning movement sequences. A different stimulus was presented in each step of the sequence and required a different movement reaction, and reward occurred at the end of the correctly performed sequence. The dopamine-like predictive reinforcement signal efficiently allowed the model to learn long sequences. By contrast, learning with an unconditional reinforcement signal required synaptic eligibility traces of longer and biologically less-plausible durations for obtaining satisfactory performance. Thus, dopamine-like neuronal signals constitute excellent teaching signals for learning sequential behavior.

Living with lethal PIP3 levels: viability of flies lacking PTEN restored by a PH domain mutation in Akt/PKB.

Abstract: The phosphoinositide phosphatase PTEN is mutated in many human cancers. Although the role of PTEN has been studied extensively, the relative contributions of its numerous potential downstream effectors to deregulated growth and tumorigenesis remain uncertain. We provide genetic evidence in Drosophila melanogaster for the paramount importance of the protein kinase Akt [also called protein kinase B (PKB)] in mediating the effects of increased phosphatidylinositol 3,4,5-trisphosphate (PIP3) concentrations that are caused by the loss of PTEN function. A mutation in the pleckstrin homology (PH) domain of Akt that reduces its affinity for PIP3 sufficed to rescue the lethality of flies devoid of PTEN activity. Thus, Akt appears to be the only critical target activated by increased PIP3 concentrations in Drosophila.

The regenerative capacity of the zebrafish heart is dependent on TGFβ signaling

Abstract: Mammals respond to a myocardial infarction by irreversible scar formation. By contrast, zebrafish are able to resolve the scar and to regenerate functional cardiac muscle. It is not known how opposing cellular responses of fibrosis and new myocardium formation are spatially and temporally coordinated during heart regeneration in zebrafish. Here, we report that the balance between the reparative and regenerative processes is achieved through Smad3-dependent TGFβ signaling. The type I receptor alk5b (tgfbr1b) is expressed in both fibrotic and cardiac cells of the injured heart. TGFβ ligands are locally induced following cryoinjury and activate the signaling pathway both in the infarct area and in cardiomyocytes in the vicinity of the trauma zone. Inhibition of the relevant type I receptors with the specific chemical inhibitor SB431542 qualitatively altered the infarct tissue and completely abolished heart regeneration. We show that transient scar formation is an essential step to maintain robustness of the damaged ventricular wall prior to cardiomyocyte replacement. Taking advantage of the reversible action of the inhibitor, we dissected the multifunctional role of TGFβ signaling into three crucial processes: collagen-rich scar deposition, Tenascin C-associated tissue remodeling at the infarct-myocardium interface, and cardiomyocyte proliferation. Thus, TGFβ signaling orchestrates the beneficial interplay between scar-based repair and cardiomyocyte-based regeneration to achieve complete heart regeneration.

Articular cartilage: from formation to tissue engineering

Abstract: Hyaline cartilage is the nonlinear, inhomogeneous, anisotropic, poro-viscoelastic connective tissue that serves as friction-reducing and load-bearing cushion in synovial joints and is vital for mammalian skeletal movements. Due to its avascular nature, low cell density, low proliferative activity and the tendency of chondrocytes to de-differentiate, cartilage cannot regenerate after injury, wear and tear, or degeneration through common diseases such as osteoarthritis. Therefore severe damage usually requires surgical intervention. Current clinical strategies to generate new tissue include debridement, microfracture, autologous chondrocyte transplantation, and mosaicplasty. While articular cartilage was predicted to be one of the first tissues to be successfully engineered, it proved to be challenging to reproduce the complex architecture and biomechanical properties of the native tissue. Despite significant research efforts, only a limited number of studies have evolved up to the clinical trial stage. This review article summarizes the current state of cartilage tissue engineering in the context of relevant biological aspects, such as the formation and growth of hyaline cartilage, its composition, structure and biomechanical properties. Special attention is given to materials development, scaffold designs, fabrication methods, and template-cell interactions, which are of great importance to the structure and functionality of the engineered tissue.