Search

We introduce DynaCLR, a self-supervised framework for modeling cell dynamics via contrastive learning of representations of time-lapse datasets. Live cell imaging of cells and organelles is widely used to analyze cellular responses to perturbations. Human annotation of dynamic cell states captured by time-lapse perturbation datasets is laborious and prone to bias. DynaCLR integrates single-cell tracking with time-aware contrastive learning to map images of cells at neighboring time points to neighboring embeddings. Mapping the morphological dynamics of cells to a temporally regularized embedding space makes the annotation, classification, clustering, or interpretation of the cell states more quantitative and efficient. We illustrate the features and applications of DynaCLR with the following experiments: analyzing the kinetics of viral infection in human cells, detecting transient changes in cell morphology due to cell division, and mapping the dynamics of organelles due to viral infection. Models trained with DynaCLR consistently achieve $>95\%$ accuracy for infection state classification, enable the detection of transient cell states and reliably embed unseen experiments. DynaCLR provides a flexible framework for comparative analysis of cell state dynamics due to perturbations, such as infection, gene knockouts, and drugs. We provide PyTorch-based implementations of the model training and inference pipeline (https://github.com/mehta-lab/viscy) and a user interface (https://github.com/czbiohub-sf/napari-iohub) for the visualization and annotation of trajectories of cells in the real space and the embedding space., Comment: 20 pages, 6 figures, 3 appendix figures, 4 videos (ancillary files)

Frameworks and DSLs auto-generating code have traditionally relied on human experts developing them to have in place rigorous methods to assure the legality of the applied code transformations. Machine Learning (ML) is gaining wider adoption as a means to auto-generate code optimised for the hardware target. However, ML solutions, and in particular black-box DNNs, provide no such guarantees on legality. In this paper we propose a library, Tadashi, which leverages the polyhedral model to empower researchers seeking to curate datasets crucial for applying ML in code-generation. Tadashi provides the ability to reliably and practically check the legality of candidate transformations on polyhedral schedules applied on a baseline reference code. We provide a proof that our library guarantees the legality of generated transformations, and demonstrate its lightweight practical cost. Tadashi is available at https://github.com/vatai/tadashi/., Comment: Submitted to CGO

Defects in semiconductors have in recent years been revealed to have interesting properties in the venture towards quantum technologies. In this regard, silicon carbide has shown great promise as a host for quantum defects. In particular, the ultra-bright AB photoluminescence lines in 4H-SiC are observable at room temperature and have been proposed as a single-photon quantum emitter. These lines have been previously studied and assigned to the carbon antisite-vacancy pair (CAV). In this paper, we report on new measurements of the AB-lines' temperature dependence, and carry out an in-depth computational study on the optical properties of the CAV defect. We find that the CAV defect has the potential to exhibit several different zero-phonon luminescences with emissions in the near-infrared telecom band, in its neutral and positive charge states. However, our measurements show that the AB-lines only consist of three non-thermally activated lines instead of the previously reported four lines, meanwhile our calculations on the CAV defect are unable to find optical transitions in full agreement with the AB-line assignment. In the light of our results, the identification of the AB-lines and the associated room temperature emission require further study., Comment: 14 main pages, 7 main figures, 2 tables, one appendix with software details, and supplementary with 2 figures and 2 tables

The size and complexity of software applications is increasing at an accelerating pace. Source code repositories (along with their dependencies) require vast amounts of labor to keep them tested, maintained, and up to date. As the discipline now begins to also incorporate automatically generated programs, automation in testing and tuning is required to keep up with the pace - let alone reduce the present level of complexity. While machine learning has been used to understand and generate code in various contexts, machine learning models themselves are trained almost exclusively on static code without inputs, traces, or other execution time information. This lack of training data limits the ability of these models to understand real-world problems in software. In this work we show that inputs, like code, can be generated automatically at scale. Our generated inputs are stateful, and appear to faithfully reproduce the arbitrary data structures and system calls required to rerun a program function. By building our tool within the compiler, it both can be applied to arbitrary programming languages and architectures and can leverage static analysis and transformations for improved performance. Our approach is able to produce valid inputs, including initial memory states, for 90% of the ComPile dataset modules we explored, for a total of 21.4 million executable functions. Further, we find that a single generated input results in an average block coverage of 37%, whereas guided generation of five inputs improves it to 45%.

Quantum communication promises unprecedented communication capabilities enabled by the transmission of quantum states of light. However, current implementations face severe limitations in communication distance due to photon loss. Silicon carbide (SiC) defects have emerged as a promising quantum device platform, offering strong optical transitions, long spin coherence lifetimes and the opportunity for integration with semiconductor devices. Some defects with optical transitions in the telecom range have been identified, allowing to interface with fiber networks without the need for wavelength conversion. These unique properties make SiC an attractive platform for the implementation of quantum nodes for quantum communication networks. We provide an overview of the most prominent defects in SiC and their implementation in spin-photon interfaces. Furthermore, we model a memory-enhanced quantum communication protocol in order to extract the parameters required to surpass a direct point-to-point link performance. Based on these insights, we summarize the key steps required towards the deployment of SiC devices in large-scale quantum communication networks., Comment: 20 pages, 8 figures

Spin-active quantum emitters have emerged as a leading platform for quantum technologies. However, one of their major limitations is the large spread in optical emission frequencies, which typically extends over tens of GHz. Here, we investigate single V4+ vanadium centres in 4H-SiC, which feature telecom-wavelength emission and a coherent S=1/2 spin state. We perform spectroscopy on single emitters and report the observation of spin-dependent optical transitions, a key requirement for spin-photon interfaces. By engineering the isotopic composition of the SiC matrix, we reduce the inhomogeneous spectral distribution of different emitters down to 100 MHz, significantly smaller than any other single quantum emitter. Additionally, we tailor the dopant concentration to stabilise the telecom-wavelength V4+ charge state, thereby extending its lifetime by at least two orders of magnitude. These results bolster the prospects for single V emitters in SiC as material nodes in scalable telecom quantum networks., Comment: 29 pages, 20 figures, 2 tables

The origin of life is shrouded in mystery, with few surviving clues, obscured by evolutionary competition. Previous reviews have touched on the complementary approaches of top-down and bottom-up synthetic biology to augment our understanding of living systems. Here we point out the synergies between these fields, especially between bottom-up synthetic biology and origin of life research. We explore recent progress made in artificial cell compartmentation in line with the crowded cell, its metabolism, as well as cycles of growth and division, and how those efforts are starting to be combined. Though the complexity of current life is among its most striking characteristics, none of life's essential features require it, and they are unlikely to have emerged thus complex from the beginning. Rather than recovering the one true origin lost in time, current research converges towards reproducing the emergence of minimal life, by teasing out how complexity and evolution may arise from a set of essential components., Comment: 29 pages, 2 figures, 1 table

While parallelism remains the main source of performance, architectural implementations and programming models change with each new hardware generation, often leading to costly application re-engineering. Most tools for performance portability require manual and costly application porting to yet another programming model. We propose an alternative approach that automatically translates programs written in one programming model (CUDA), into another (CPU threads) based on Polygeist/MLIR. Our approach includes a representation of parallel constructs that allows conventional compiler transformations to apply transparently and without modification and enables parallelism-specific optimizations. We evaluate our framework by transpiling and optimizing the CUDA Rodinia benchmark suite for a multi-core CPU and achieve a 76% geomean speedup over handwritten OpenMP code. Further, we show how CUDA kernels from PyTorch can efficiently run and scale on the CPU-only Supercomputer Fugaku without user intervention. Our PyTorch compatibility layer making use of transpiled CUDA PyTorch kernels outperforms the PyTorch CPU native backend by 2.7$\times$.

The negatively charged silicon vacancy ($\mathrm{V_{Si}^-}$) in silicon carbide is a well-studied point defect for quantum applications. At the same time, a closer inspection of ensemble photoluminescence and electron paramagnetic resonance measurements reveals an abundance of related but so far unidentified signals. In this study, we search for defects in 4H-SiC that explain the above magneto-optical signals in a defect database generated by Automatic Defect Analysis and Qualification (ADAQ) workflows. This search reveals only one class of atomic structures that exhibit silicon-vacancy-like properties in the data: a carbon antisite ($\mathrm{C_{Si}}$) within sub-nanometer distances from the silicon vacancy only slightly alters the latter without affecting the charge or spin state. Such a perturbation is energetically bound. We consider the formation of $\mathrm{V_{Si}^-+C_{Si}}$ up to 2 nm distance and report their zero phonon lines and zero field splitting values. In addition, we perform high-resolution photoluminescence experiments in the silicon vacancy region and find an abundance of lines. Comparing our computational and experimental results, several configurations show great agreement. Our work demonstrates the effectiveness of a database with high-throughput results in the search for defects in quantum applications.

Spin-active color centers in solids show good performance for quantum technologies. Several transition-metal defects in SiC offer compatibility with telecom and semiconductor industries. However, whether their strong spin-orbit coupling degrades their spin lifetimes is not clear. We show that a combination of a crystal-field with axial symmetry and spin-orbit coupling leads to a suppression of spin-lattice and spin-spin interactions, resulting in remarkably slow spin relaxation. Our optical measurements on an ensemble of Mo impurities in SiC show a spin lifetime $T_1$ of $2,4$ s at $2$ K.

We study a generalized discrete-time multi-type Wright-Fisher population process. The mean-field dynamics of the stochastic process is induced by a general replicator difference equation. We prove several results regarding the asymptotic behavior of the model, focusing on the impact of the mean-field dynamics on it. One of the results is a limit theorem that describes sufficient conditions for an almost certain path to extinction, first eliminating the type which is the least fit at the mean-field equilibrium. The effect is explained by the metastability of the stochastic system, which under the conditions of the theorem spends almost all time before the extinction event in a neighborhood of the equilibrium. In addition, to limit theorems, we propose a variation of Fisher's maximization principle, fundamental theorem of natural selection, for a completely general deterministic replicator dynamics and study implications of the deterministic maximization principle for the stochastic model.

Colour centres with long-lived spins are established platforms for quantum sensing and quantum information applications. Colour centres exist in different charge states, each of them with distinct optical and spin properties. Application to quantum technology requires the capability to access and stabilize charge states for each specific task. Here, we investigate charge state manipulation of individual silicon vacancies in silicon carbide, a system which has recently shown a unique combination of long spin coherence time and ultrastable spin-selective optical transitions. In particular, we demonstrate charge state switching through the bias applied to the colour centre in an integrated silicon carbide opto-electronic device. We show that the electronic environment defined by the doping profile and the distribution of other defects in the device plays a key role for charge state control. Our experimental results and numerical modeling evidence that control of these complex interactions can, under certain conditions, enhance the photon emission rate. These findings open the way for deterministic control over the charge state of spin-active colour centres for quantum technology and provide novel techniques for monitoring doping profiles and voltage sensing in microscopic devices.

Optically interfaced spins in the solid promise scalable quantum networks. Robust and reliable optical properties have so far been restricted to systems with inversion symmetry. Here, we release this stringent constraint by demonstrating outstanding optical and spin properties of single silicon vacancy centres in silicon carbide. Despite the lack of inversion symmetry, the system's particular wave function symmetry decouples its optical properties from magnetic and electric fields, as well as from local strain. This provides a high-fidelity spin-to-photon interface with exceptionally stable and narrow optical transitions, low inhomogeneous broadening, and a large fraction of resonantly emitted photons. Further, the weak spin-phonon coupling results in electron spin coherence times comparable with nitrogen-vacancy centres in diamond. This allows us to demonstrate coherent hyperfine coupling to single nuclear spins, which can be exploited as qubit memories. Our findings promise quantum network applications using integrated semiconductor-based spin-to-photon interfaces., Comment: Main text: 16 pages, 4 figures. Supplementary information: 16 pages, 4 figures

Robots that solve complex tasks in environments too dangerous for humans to enter are desperately needed, e.g. for search and rescue applications. We describe our mobile manipulation robot Momaro, with which we participated successfully in the DARPA Robotics Challenge. It features a unique locomotion design with four legs ending in steerable wheels, which allows it both to drive omnidirectionally and to step over obstacles or climb. Furthermore, we present advanced communication and teleoperation approaches, which include immersive 3D visualization, and 6D tracking of operator head and arm motions. The proposed system is evaluated in the DARPA Robotics Challenge, the DLR SpaceBot Cup Qualification and lab experiments. We also discuss the lessons learned from the competitions.

We investigate the quenching of the photoluminescence (PL) from the divacancy defect in 4H-SiC consisting of a nearest-neighbour silicon and carbon vacancies. The quenching occurs only when the PL is excited below certain photon energies (thresholds), which differ for the four different inequivalent divacancy configurations in 4H-SiC. Refined theoretical ab initio calculation for the charge-transfer levels of the divacancy show very good agreement between the position of the (0/-) level with respect to the conduction band for each divacancy configurations and the corresponding experimentally observed threshold, allowing us to associate the PL decay with conversion of the divacancy from neutral to negative charge state due to capture of electrons photoionized from other defects (traps) by the excitation. Electron paramagnetic resonance measurements are conducted in dark and under excitation similar to that used in the PL experiments and shed light on the possible origin of traps in the different samples. A simple model built on this concept agrees well with the experimentally-observed decay curves., Comment: 28 pages, 6 figures

Color centers in wide-bandgap semiconductors are attractive systems for quantum technologies since they can combine long-coherent electronic spin and bright optical properties. Several suitable centers have been identified, most famously the nitrogen-vacancy defect in diamond. However, integration in communication technology is hindered by the fact that their optical transitions lie outside telecom wavelength bands. Several transition-metal impurities in silicon carbide do emit at and near telecom wavelengths, but knowledge about their spin and optical properties is incomplete. We present all-optical identification and coherent control of molybdenum-impurity spins in silicon carbide with transitions at near-infrared wavelengths. Our results identify spin $S=1/2$ for both the electronic ground and excited state, with highly anisotropic spin properties that we apply for implementing optical control of ground-state spin coherence. Our results show optical lifetimes of $\sim$60 ns and inhomogeneous spin dephasing times of $\sim$0.3 $\mu$s, establishing relevance for quantum spin-photon interfacing., Comment: Updated version with minor correction, full Supplementary Information included

The controlled generation and manipulation of atom-like defects in solids has a wide range of applications in quantum technology. Although various defect centres have displayed promise as either quantum sensors, single photon emitters or light-matter interfaces, the search for an ideal defect with multi-functional ability remains open. In this spirit, we investigate here the optical and spin properties of the V1 defect centre, one of the silicon vacancy defects in the 4H polytype of silicon carbide (SiC). The V1 centre in 4H-SiC features two well-distinguishable sharp optical transitions and a unique S=3/2 electronic spin, which holds promise to implement a robust spin-photon interface. Here, we investigate the V1 defect at low temperatures using optical excitation and magnetic resonance techniques. The measurements, which are performed on ensemble, as well as on single centres, prove that this centre combines coherent optical emission, with up to 40% of the radiation emitted into the zero-phonon line (ZPL), a strong optical spin signal and long spin coherence time. These results single out the V1 defect in SiC as a promising system for spin-based quantum technologies.