IST Austria: PubRep (Institute of Science and Technology)
Not a member yet
6140 research outputs found
Sort by
LNCS
No full textDespite the advances in probabilistic model checking, the scalability of the verification methods remains limited. In particular, the state space often becomes extremely large when instantiating parameterized Markov decision processes (MDPs) even with moderate values. Synthesizing policies for such huge MDPs is beyond the reach of available tools. We propose a learning-based approach to obtain a reasonable policy for such huge MDPs.
The idea is to generalize optimal policies obtained by model-checking small instances to larger ones using decision-tree learning. Consequently, our method bypasses the need for explicit state-space exploration of large models, providing a practical solution to the state-space explosion problem. We demonstrate the efficacy of our approach by performing extensive experimentation on the relevant models from the quantitative verification benchmark set. The experimental results indicate that our policies perform well, even when the size of the model is orders of magnitude beyond the reach of state-of-the-art analysis tools
Remodeling of lipid-foam prototissues by network-wide tension fluctuations induced by active particles
No full textRecent advances in the field of bottom-up synthetic biology have led to the development of synthetic cells that mimic some features of real cells, such as division, protein synthesis, or DNA replication. Larger assemblies of synthetic cells may be used to form prototissues. However, existing prototissues are limited by their relatively small lateral dimensions or their lack of remodeling ability. Here, we introduce a lipid-based tissue mimetic that can be easily prepared and functionalized, consisting of a millimeter-sized “lipid-foam” with individual micrometer-sized compartments bound by lipid bilayers. We characterize the structural and mechanical properties of the lipid-foam tissue mimetic, and we demonstrate self-healing capabilities enabled by the fluidity of the lipid bilayers. Upon inclusion of bacteria in the tissue compartments, we observe that the tissue mimetic exhibits network-wide tension fluctuations driven by membrane tension generation by the swimming bacteria. Active tension fluctuations facilitate the fluidization and reorganization of the prototissue, providing a versatile platform for understanding and mimicking biological tissues
LNCS
No full textIn reconfiguration, we are given two solutions to a graph problem, such as Vertex Cover or Dominating Set, with each solution represented by a placement of tokens on vertices of the graph. Our task is to reconfigure one into the other using small steps while ensuring the intermediate configurations of tokens are also valid solutions. The two commonly studied settings are Token Jumping and Token Sliding, which allows moving a single token to an arbitrary or an adjacent vertex, respectively.
We introduce new rules that generalize Token Jumping, parameterized by the number of tokens allowed to move at once and by the maximum distance of each move. Our main contribution is identifying minimal rules that allow reconfiguring any possible given solution into any other for Independent Set, Vertex Cover, and Dominating Set. For each minimal rule, we also provide an efficient algorithm that finds a corresponding reconfiguration sequence.
We further focus on the rule that allows each token to move to an adjacent vertex in a single step. This natural variant turns out to be the minimal rule that guarantees reconfigurability for Vertex Cover. We determine the computational complexity of deciding whether a (shortest) reconfiguration sequence exists under this rule for the three studied problems. While reachability for Vertex Cover is shown to be in P, finding a shortest sequence is shown to be NP-complete. For Independent Set and Dominating Set, even reachability is shown to be PSPACE-complete
FakET: Simulating cryo-electron tomograms with neural style transfer
No full textIn cryo-electron microscopy, accurate particle localization and classification are imperative. Recent deep learning solutions, though successful, require extensive training datasets. The protracted generation time of physics-based models, often employed to produce these datasets, limits their broad applicability. We introduce FakET, a method based on neural style transfer, capable of simulating the forward operator of any cryo transmission electron microscope. It can be used to adapt a synthetic training dataset according to reference data producing high-quality simulated micrographs or tilt-series. To assess the quality of our generated data, we used it to train a state-of-the-art localization and classification architecture and compared its performance with a counterpart trained on benchmark data. Remarkably, our technique matches the performance, boosts data generation speed 750x, uses 33x less memory, and scales well to typical transmission electron microscope detector sizes. It leverages GPU acceleration and parallel processing. The source code is available at https://github.com/paloha/faket/
Domain-wall ferroelectric polarons in a two-dimensional rotor lattice model
No full textWe demonstrate the formation of ferroelectric domain-wall polarons in a minimal two-dimensional lattice model of electrons interacting with rotating dipoles. Along the domain wall, the rotors polarize in opposite directions, causing the electron to localize along a particular lattice direction. The rotor-electron coupling is identified as the origin of a structural instability in the crystal that leads to the domain-wall formation via a symmetry-breaking process. Our results provide the first theoretical description of ferroelectric polarons, as discussed in the context of soft semiconductors
Learning reshapes the hippocampal representation hierarchy
No full textA key feature of biological and artificial neural networks is the progressive refinement of their neural representations with experience. In neuroscience, this fact has inspired several recent studies in sensory and motor systems. However, less is known about how higher associational cortical areas, such as the hippocampus, modify representations throughout the learning of complex tasks. Here, we focus on associative learning, a process that requires forming a connection between the representations of different variables for appropriate behavioral response. We trained rats in a space-context associative task and monitored hippocampal neural activity throughout the entire learning period, over several days. This allowed us to assess changes in the representations of context, movement direction, and position, as well as their relationship to behavior. We identified a hierarchical representational structure in the encoding of these three task variables that was preserved throughout learning. Nevertheless, we also observed changes at the lower levels of the hierarchy where context was encoded. These changes were local in neural activity space and restricted to physical positions where context identification was necessary for correct decision-making, supporting better context decoding and contextual code compression. Our results demonstrate that the hippocampal code not only accommodates hierarchical relationships between different variables but also enables efficient learning through minimal changes in neural activity space. Beyond the hippocampus, our work reveals a representation learning mechanism that might be implemented in other biological and artificial networks performing similar tasks
Sustainable development key to limiting climate change-driven wildfire damages
No full textClimate change is causing wildfires to become more frequent and intense. While predicting burned areas using bioclimatic and anthropogenic factors is an active research area, few studies have examined what drives the economic damages of wildfires. Our study aims to fill this gap by analyzing key factors influencing global economic wildfire damages and projecting future damages under three shared socioeconomic pathways (SSPs). We apply regression analyses to identify significant predictors of economic wildfire damages at country levels and use the fitted model to project future damages under SSP126, SSP245, and SSP370. Results show that the human vulnerability index (HVI), reflecting socioeconomic conditions, is the strongest predictor of historical wildfire damages, followed by water vapor pressure deficit during the fire season and population density around forested areas. We found high population density to be associated with lower damages. These findings contrast with studies of burned areas, where climate factors are more dominant. Our model projects that by 2070, average global economic wildfire damages will be three times higher under SSP370 than SSP126. Our model also shows that following SSP126 not only reduces wildfire damages but also lessens the inequalities in damage distribution across countries. This pathway’s dual focus on equitable socioeconomic progress and climate action potentially enhances a country’s resilience that helps mitigate wildfire damages. Our analyses also indicate that strong socioeconomic development can offset wildfire damages associated with climate hazards, although this is less certain under SSP370. SSP126’s integrated approach improves both socioeconomic conditions and limits global warming, providing substantial benefits to less developed countries while still reducing damages in developed nations, despite their already low HVI scores. Our work complements existing research on burned areas and underscores the importance of sustainable development and international collaboration in reducing the economic damages of wildfires
Dataset for "Bumps on the Road: The Way to Clean Relaxation Dispersion in the Solid State"
No full textKiwa is a membrane-embedded defense supercomplex activated at phage attachment sites
No full textBacteria and archaea deploy diverse antiviral defense systems, many of which remain mechanistically uncharacterized. Here, we characterize Kiwa, a widespread two-component system composed of the transmembrane sensor KwaA and the DNA-binding effector KwaB. Cryogenic electron microscopy (cryo-EM) analysis reveals that KwaA and KwaB assemble into a large, membrane-associated supercomplex. Upon phage binding, KwaA senses infection at the membrane, leading to KwaB binding of ejected phage DNA and inhibition of replication and late transcription, without inducing host cell death. Although KwaB can bind DNA independently, its antiviral activity requires association with KwaA, suggesting spatial or conformational regulation. We show that the phage-encoded DNA-mimic protein Gam directly binds and inhibits KwaB but that co-expression with the Gam-targeted RecBCD system restores protection by Kiwa. Our findings support a model in which Kiwa coordinates membrane-associated detection of phage infection with downstream DNA binding by its effector, forming a spatially coordinated antiviral mechanism
Catalogue of criteria for assessing the funding eligibility of Open Science infrastructures
No full textThis criteria catalogue and the accompanying assessment questions were developed by a working group of KEMÖ (Kooperation E-Medien Österreich, the Austrian Academic Library Consortium). They are intended to support research institutions and organisations in the evaluation of Open Science Infrastructures. The 20 criteria outlined in the catalogue provide a structured basis for making informed decisions regarding the financial support of these infrastructures.
The assessment questions are intended to be completed by Open Science Infrastructures and can be shared with them accordingly