Search

We investigate the directional characteristics of photon statistics in dimers of quantum emitters. For their analysis, we construct a two-point second-order correlation function that allows us to find a new mechanism for photon anticorrelation, termed as geometric antibunching. This phenomenon is completely agnostic to the quantum state of the emitters and emerges from quantum interference effects due to the indistinguishability of different two-photon optical pathways. Finally, we explore its occurrence in emitters placed in the vicinity of a flat substrate and a nanosphere, demonstrating its tunnability through the different material and geometric parameters of these structures., Comment: 11 pages, 6 figures

Objective:Develop and validate an algorithm for analyzing the layout of PDF clinical documents to improve the performance of downstream natural language processing tasks. Materials and Methods: We designed an algorithm to process clinical PDF documents and extract only clinically relevant text. The algorithm consists of several steps: initial text extraction using a PDF parser, followed by classification into categories such as body text, left notes, and footers using a Transformer deep neural network architecture, and finally an aggregation step to compile the lines of a given label in the text. We evaluated the technical performance of the body text extraction algorithm by applying it to a random sample of documents that were annotated. Medical performance was evaluated by examining the extraction of medical concepts of interest from the text in their respective sections. Finally, we tested an end-to-end system on a medical use case of automatic detection of acute infection described in the hospital report. Results:Our algorithm achieved per-line precision, recall, and F1 score of 98.4, 97.0, and 97.7, respectively, for body line extraction. The precision, recall, and F1 score per document for the acute infection detection algorithm were 82.54 (95CI 72.86-91.60), 85.24 (95CI 76.61-93.70), 83.87 (95CI 76, 92-90.08) with exploitation of the results of the advanced body extraction algorithm, respectively. Conclusion:We have developed and validated a system for extracting body text from clinical documents in PDF format by identifying their layout. We were able to demonstrate that this preprocessing allowed us to obtain better performances for a common downstream task, i.e., the extraction of medical concepts in their respective sections, thus proving the interest of this method on a clinical use case., Comment: 22 pages, 5 figures

Transformers' compute-intensive operations pose enormous challenges for their deployment in resource-constrained EdgeAI / tinyML devices. As an established neural network compression technique, quantization reduces the hardware computational and memory resources. In particular, fixed-point quantization is desirable to ease the computations using lightweight blocks, like adders and multipliers, of the underlying hardware. However, deploying fully-quantized Transformers on existing general-purpose hardware, generic AI accelerators, or specialized architectures for Transformers with floating-point units might be infeasible and/or inefficient. Towards this, we propose SwiftTron, an efficient specialized hardware accelerator designed for Quantized Transformers. SwiftTron supports the execution of different types of Transformers' operations (like Attention, Softmax, GELU, and Layer Normalization) and accounts for diverse scaling factors to perform correct computations. We synthesize the complete SwiftTron architecture in a $65$ nm CMOS technology with the ASIC design flow. Our Accelerator executes the RoBERTa-base model in 1.83 ns, while consuming 33.64 mW power, and occupying an area of 273 mm^2. To ease the reproducibility, the RTL of our SwiftTron architecture is released at https://github.com/albertomarchisio/SwiftTron., Comment: To appear at the 2023 International Joint Conference on Neural Networks (IJCNN), Queensland, Australia, June 2023

The objective of this study is to address the critical issue of de-identification of clinical reports in order to allow access to data for research purposes, while ensuring patient privacy. The study highlights the difficulties faced in sharing tools and resources in this domain and presents the experience of the Greater Paris University Hospitals (AP-HP) in implementing a systematic pseudonymization of text documents from its Clinical Data Warehouse. We annotated a corpus of clinical documents according to 12 types of identifying entities, and built a hybrid system, merging the results of a deep learning model as well as manual rules. Our results show an overall performance of 0.99 of F1-score. We discuss implementation choices and present experiments to better understand the effort involved in such a task, including dataset size, document types, language models, or rule addition. We share guidelines and code under a 3-Clause BSD license.

Unstructured proteins can modulate cellular responses to environmental conditions by undergoing coil-globule transitions and phase separation. However, the molecular mechanisms of these phenomena still need to be fully understood. Here, we use Monte Carlo calculations of a coarse-grained model incorporating water's effects on the system's free energy. Following previous studies, we model an unstructured protein as a polymer chain. Because we are interested in investigating how it responds to thermodynamic changes near a hydrophobic surface under different conditions, we choose an entirely hydrophobic sequence to maximize the interaction with the interface. We show that a slit pore confinement without top-down symmetry enhances the unfolding and adsorption of the chain in both random coil and globular states. Moreover, we demonstrate that the hydration water modulates this behavior depending on the thermodynamic parameters. Our findings provide insights into how homopolymers, and possibly unstructured proteins, can sense and adjust to external stimuli such as nanointerfaces or stresses., Comment: 25 pages, 5 figures,8 supplementary movies (mp4)

We present a joint experimental and theoretical study of non-sequential double ionization (NSDI) in argon driven by a 3100-nm laser source. The correlated photoelectron momentum distribution (PMD) shows a strong dependence on the pulse duration, and the evolution of the PMD can be explained by an envelope-induced intensity effect. Determined by the time difference between tunneling and rescattering, the laser vector potential at the ionization time of the bound electron will be influenced by the pulse duration, leading to different drift momenta. Such a mechanism is extracted through a classical trajectory Monte Carlo-based model and it can be further confirmed by quantum mechanical simulations. This work sheds light on the importance of the pulse duration in NSDI and improves our understanding of the strong field tunnel-recollision dynamics under mid-IR laser fields., Comment: 19 pages, 9 figures, submitted to NJP, comments are welcomed

Background Clinical studies using real-world data may benefit from exploiting clinical reports, a particularly rich albeit unstructured medium. To that end, natural language processing can extract relevant information. Methods based on transfer learning using pre-trained language models have achieved state-of-the-art results in most NLP applications; however, publicly available models lack exposure to speciality-languages, especially in the medical field. Objective We aimed to evaluate the impact of adapting a language model to French clinical reports on downstream medical NLP tasks. Methods We leveraged a corpus of 21M clinical reports collected from August 2017 to July 2021 at the Greater Paris University Hospitals (APHP) to produce two CamemBERT architectures on speciality language: one retrained from scratch and the other using CamemBERT as its initialisation. We used two French annotated medical datasets to compare our language models to the original CamemBERT network, evaluating the statistical significance of improvement with the Wilcoxon test. Results Our models pretrained on clinical reports increased the average F1-score on APMed (an APHP-specific task) by 3 percentage points to 91%, a statistically significant improvement. They also achieved performance comparable to the original CamemBERT on QUAERO. These results hold true for the fine-tuned and from-scratch versions alike, starting from very few pre-training samples. Conclusions We confirm previous literature showing that adapting generalist pre-train language models such as CamenBERT on speciality corpora improves their performance for downstream clinical NLP tasks. Our results suggest that retraining from scratch does not induce a statistically significant performance gain compared to fine-tuning.

Recent advances in neural text-to-speech research have been dominated by two-stage pipelines utilizing low-level intermediate speech representation such as mel-spectrograms. However, such predetermined features are fundamentally limited, because they do not allow to exploit the full potential of a data-driven approach through learning hidden representations. For this reason, several end-to-end methods have been proposed. However, such models are harder to train and require a large number of high-quality recordings with transcriptions. Here, we propose WavThruVec - a two-stage architecture that resolves the bottleneck by using high-dimensional Wav2Vec 2.0 embeddings as intermediate speech representation. Since these hidden activations provide high-level linguistic features, they are more robust to noise. That allows us to utilize annotated speech datasets of a lower quality to train the first-stage module. At the same time, the second-stage component can be trained on large-scale untranscribed audio corpora, as Wav2Vec 2.0 embeddings are already time-aligned. This results in an increased generalization capability to out-of-vocabulary words, as well as to a better generalization to unseen speakers. We show that the proposed model not only matches the quality of state-of-the-art neural models, but also presents useful properties enabling tasks like voice conversion or zero-shot synthesis., Comment: Accepted to INTERSPEECH 2022. Audio samples are available at: https://charactr-platform.github.io/WavThruVec/

Novel text-to-speech systems can generate entirely new voices that were not seen during training. However, it remains a difficult task to efficiently create personalized voices from a high-dimensional speaker space. In this work, we use speaker embeddings from a state-of-the-art speaker verification model (SpeakerNet) trained on thousands of speakers to condition a TTS model. We employ a human sampling paradigm to explore this speaker latent space. We show that users can create voices that fit well to photos of faces, art portraits, and cartoons. We recruit online participants to collectively manipulate the voice of a speaking face. We show that (1) a separate group of human raters confirms that the created voices match the faces, (2) speaker gender apparent from the face is well-recovered in the voice, and (3) people are consistently moving towards the real voice prototype for the given face. Our results demonstrate that this technology can be applied in a wide number of applications including character voice development in audiobooks and games, personalized speech assistants, and individual voices for people with speech impairment., Comment: Accepted to Interspeech'22. Audio and video samples are available at: https://polvanrijn.github.io/VoiceMe/

Time-resolved XUV-IR photoion mass spectroscopy of naphthalene conducted with broadband, as well as with wavelength-selected narrowband XUV pulses reveals a rising probability of fragmentation characterized by a lifetime of $92\pm4$~fs. This lifetime is independent of the XUV excitation wavelength and is the same for all low appearance energy fragments recorded in the experiment. Analysis of the experimental data in conjunction with a statistical multi-state vibronic model suggests that the experimental signals track vibrational energy redistribution on the potential energy surface of the ground state cation. In particular, populations of the out-of-plane ring twist and the out-of-plane wave bending modes could be responsible for opening new IR absorption channels leading to enhanced fragmentation., Comment: Submitted to Daniel Neumark's Festschrift

We propose and investigate a hidden Markov model (HMM) for the analysis of dependent, aggregated, superimposed two-state signal recordings. A major motivation for this work is that often these signals cannot be observed individually but only their superposition. Among others, such models are in high demand for the understanding of cross-talk between ion channels, where each single channel cannot be measured separately. As an essential building block, we introduce a parameterized vector norm dependent Markov chain model and characterize it in terms of permutation invariance as well as conditional independence. This building block leads to a hidden Markov chain sum process which can be used for analyzing the dependence structure of superimposed two-state signal observations within an HMM. Notably, the model parameters of the vector norm dependent Markov chain are uniquely determined by the parameters of the sum process and are therefore identifiable. We provide algorithms to estimate the parameters, discuss model selection and apply our methodology to real-world ion channel data from the heart muscle, where we show competitive gating., Comment: 49 pages, 11 figures. An R package can be found at: https://github.com/ljvanegas/VND

We reveal the assembly of monodisperse magnetite nanoparticles of sizes 5 nm, 15 nm and 25 nm from dilute water-based ferrofluids onto an amorphous magnetic template with out-of-plane anisotropy. From neutron reflectometry experiments we extract density profiles and show that the particles self-assemble into layers at the magnetic surface. The layers are extremely stable against cleaning and rinsing of the substrate. The density of the layers is determined by and increases with the remanent magnetic moment of the particles.

The Potjans-Diesmann cortical microcircuit model is a widely used model originallyimplemented in NEST. Here, we re-implemented the model using NetPyNE, a high-level Python interface to the NEURON simulator, and reproduced the findings of theoriginal publication. We also implemented a method for rescaling the network sizewhich preserves first and second order statistics, building on existing work on networktheory. The new implementation enables using more detailed neuron models with mul-ticompartment morphologies and multiple biophysically realistic channels. This opensthe model to new research, including the study of dendritic processing, the influenceof individual channel parameters, and generally multiscale interactions in the network.The rescaling method provides flexibility to increase or decrease the network size ifrequired when running these more realistic simulations. Finally, NetPyNE facilitatesmodifying or extending the model using its declarative language; optimizing modelparameters; running efficient large-scale parallelized simulations; and analyzing themodel through built-in methods, including local field potential calculation and informa-tion flow measures.

Machine learning is increasingly recognized as a promising technology in the biological, biomedical, and behavioral sciences. There can be no argument that this technique is incredibly successful in image recognition with immediate applications in diagnostics including electrophysiology, radiology, or pathology, where we have access to massive amounts of annotated data. However, machine learning often performs poorly in prognosis, especially when dealing with sparse data. This is a field where classical physics-based simulation seems to remain irreplaceable. In this review, we identify areas in the biomedical sciences where machine learning and multiscale modeling can mutually benefit from one another: Machine learning can integrate physics-based knowledge in the form of governing equations, boundary conditions, or constraints to manage ill-posted problems and robustly handle sparse and noisy data; multiscale modeling can integrate machine learning to create surrogate models, identify system dynamics and parameters, analyze sensitivities, and quantify uncertainty to bridge the scales and understand the emergence of function. With a view towards applications in the life sciences, we discuss the state of the art of combining machine learning and multiscale modeling, identify applications and opportunities, raise open questions, and address potential challenges and limitations. We anticipate that it will stimulate discussion within the community of computational mechanics and reach out to other disciplines including mathematics, statistics, computer science, artificial intelligence, biomedicine, systems biology, and precision medicine to join forces towards creating robust and efficient models for biological systems.

In the analysis of neutron scattering measurements of condensed matter structure, it normally suffices to treat the incident and scattered neutron beams as if composed of incoherent distributions of plane waves with wavevectors of different magnitudes and directions which are taken to define an instrumental resolution. However, despite the wide-ranging applicability of this conventional treatment, there are cases in which the wave function of an individual neutron in the beam must be described more accurately by a spatially localized packet, in particular with respect to its transverse extent normal to its mean direction of propagation. One such case involves the creation of orbital angular momentum (OAM) states in a neutron via interaction with a material device of a given size. It is shown in the work reported here that there exist two distinct measures of coherence of special significance and utility for describing neutron beams in scattering studies of materials in general. One measure corresponds to the coherent superposition of basis functions and their wavevectors which constitute each individual neutron packet state function whereas the other measure can be associated with an incoherent distribution of mean wavevectors of the individual neutron packets in a beam. Both the distribution of the mean wavevectors of individual packets in the beam as well as the wavevector components of the superposition of basis functions within an individual packet can contribute to the conventional notion of instrumental resolution. However, it is the transverse spatial extent of packet wavefronts alone that determines the area within which a coherent scattering process can occur in the first place. This picture is shown to be consistent with standard quantum theory. It is also demonstrated that these two measures of coherence can be distinguished from one another experimentally., Comment: 41 pages (including appendices), 21 figures, 4 tables

Fueled by breakthrough technology developments, the biological, biomedical, and behavioral sciences are now collecting more data than ever before. There is a critical need for time- and cost-efficient strategies to analyze and interpret these data to advance human health. The recent rise of machine learning as a powerful technique to integrate multimodality, multifidelity data, and reveal correlations between intertwined phenomena presents a special opportunity in this regard. However, classical machine learning techniques often ignore the fundamental laws of physics and result in ill-posed problems or non-physical solutions. Multiscale modeling is a successful strategy to integrate multiscale, multiphysics data and uncover mechanisms that explain the emergence of function. However, multiscale modeling alone often fails to efficiently combine large data sets from different sources and different levels of resolution. We show how machine learning and multiscale modeling can complement each other to create robust predictive models that integrate the underlying physics to manage ill-posed problems and explore massive design spaces. We critically review the current literature, highlight applications and opportunities, address open questions, and discuss potential challenges and limitations in four overarching topical areas: ordinary differential equations, partial differential equations, data-driven approaches, and theory-driven approaches. Towards these goals, we leverage expertise in applied mathematics, computer science, computational biology, biophysics, biomechanics, engineering mechanics, experimentation, and medicine. Our multidisciplinary perspective suggests that integrating machine learning and multiscale modeling can provide new insights into disease mechanisms, help identify new targets and treatment strategies, and inform decision making for the benefit of human health.

Deep kernel learning provides an elegant and principled framework for combining the structural properties of deep learning algorithms with the flexibility of kernel methods. By means of a deep neural network, we learn a parametrized kernel operator that can be combined with a differentiable kernel algorithm during inference. While previous work within this framework has focused on learning a single kernel for large datasets, we learn a kernel family for a variety of few-shot regression tasks. Compared to single deep kernel learning, our algorithm enables the identification of the appropriate kernel for each task during inference. As such, it is well adapted for complex task distributions in a few-shot learning setting, which we demonstrate by comparing against existing state-of-the-art algorithms using real-world, few-shot regression tasks related to the field of drug discovery.

Increasing urban concentration raises operational challenges that can benefit from integrated monitoring and decision support. Such complex systems need to leverage the full stack of analytical methods, from state estimation using multi-sensor fusion for situational awareness, to prediction and computation of optimal responses. The FASTER platform that we describe in this work, deployed at nation scale and handling 1.5 billion public transport trips a year, offers such a full stack of techniques for this large-scale, real-time problem. FASTER provides fine-grained situational awareness and real-time decision support with the objective of improving the public transport commuter experience. The methods employed range from statistical machine learning to agent-based simulation and mixed-integer optimization. In this work we present an overview of the challenges and methods involved, with details of the commuter movement prediction module, as well as a discussion of open problems.

La-filled skutterudites LaxCo4Sb12 (x : 0.25 and 0.5) have been synthesized and sintered in one step under high-pressure conditions at 3.5 GPa in a piston-cylinder hydrostatic press. The structural properties of the reaction products were characterized by synchrotron X-ray powder diffraction, clearly showing an uneven filling factor of the skutterudite phases, confirmed by transmission electron microscopy. The non-homogeneous distribution of La filling atoms is adequate to produce a significant decrease in lattice thermal conductivity, mainly due to strain field scattering of high-energy phonons. Furthermore, the lanthanum filler primarily acts as an Einstein-like vibrational mode having a strong impact on the phonon scattering. Extra-low thermal conductivity values of 2.39 W/mK and 1.30 W/mK are measured for La0.25Co4Sb12 and La0.5Co4Sb12 nominal compositions at 780 K, respectively. Besides this, lanthanum atoms have contributed to increase the charge carrier concentration in the samples. In the case of La0.25Co4Sb12, there is an enhancement of the power factor and an improvement of the thermoelectric properties.

Artificial intelligence (AI) is an extensive scientific discipline which enables computer systems to solve problems by emulating complex biological processes such as learning, reasoning and self-correction. This paper presents a comprehensive review of the application of AI techniques for improving performance of optical communication systems and networks. The use of AI-based techniques is first studied in applications related to optical transmission, ranging from the characterization and operation of network components to performance monitoring, mitigation of nonlinearities, and quality of transmission estimation. Then, applications related to optical network control and management are also reviewed, including topics like optical network planning and operation in both transport and access networks. Finally, the paper also presents a summary of opportunities and challenges in optical networking where AI is expected to play a key role in the near future.

This report targets decision-makers and influencers within education systems who contribute to policy-making at European, Member State or local level, and who stimulate and facilitate the development and uptake of new educational content and resources. According to the latest EU statistics 9.39 million animals were used in research and testing in 2017. A similar number of laboratory animals were bred for science but never actually used. Although in recent years the numbers have declined, there is a long way to go before the EU's goal of eliminating animal-use in science is reached. In the European Union, Directive 2010/63/EU on the protection of animals used for scientific purposes fully embraces the principles of the Three Rs, to Replace, Reduce and Refine the use of animals for experimental or educational purposes. The European Commission's Joint Research Centre (JRC) undertakes a number of initiatives to improve knowledge sharing on the Three Rs to encourage their uptake and implementation. In 2016 the JRC conducted a study to identify existing Three Rs knowledge sources and to understand how they are used and shared (Holley et al., 2016). One of the key findings was that Three Rs knowledge sharing could be enhanced through the development of targeted education and training initiatives and resources, and a subsequent study was conducted to map and characterise the already existing ones worldwide. A main conclusion was that, although Three Rs courses and resources are abundant, there is a need to further complement and amplify current education and training opportunities, both at the level of secondary school and higher education. Another related study undertaken focused on how to incorporate the Three Rs and related teaching material into the curriculum of schools and universities. This report is based on the knowledge and input gathered from a variety of expert sources and intends not only to provide the 'business case' for introducing the Three Rs into curricula and education programmes, but also suggests how this can be achieved using different means. It includes examples of learning scenarios and teaching resources to both inspire and facilitate action. [This work was supported in part through funds kindly made available by the European Parliament.]

The primary purpose of this investigation is to determine the effective coherent extent of the neutron wave packet transverse to its mean propagation vector k, when it is prepared in a typical instrument used to study the structure of materials in thin film form via specular reflection. There are two principal reasons for doing so. One has to do with the fundamental physical interest in the characteristics of a free neutron as a quantum object while the other is of a more practical nature, relating to the understanding of how to interpret elastic scattering data when the neutron is employed as a probe of condensed matter structure on an atomic or nanometer scale. Knowing such a basic physical characteristic as the neutron's effective transverse coherence can dictate how to properly analyze specular reflectivity data obtained for material film structures possessing some amount of in-plane inhomogeneity. In this study we describe a means of measuring the effective transverse coherence length of the neutron wave packet by specular reflection from a series of diffraction gratings of different spacings. Complementary non-specular measurements of the widths of grating reflections were also performed which corroborate the specular results. (Part I principally describes measurements interpreted according to the theoretical picture presented in Part II.) Each grating was fabricated by lift-off photo-lithography patterning of a nickel film (approximately 1000 Angstroms thick) formed by physical vapor deposition on a flat silicon crystal surface. The grating periods ranged from 10 microns (5 microns Ni stripe, 5 microns intervening space) to several hundred microns. The transverse coherence length, modeled as the width of the wave packet, was determined from an analysis of the specular reflectivity curves of the set of gratings., Comment: 9 January 2014 Re: AJ10970 Dear Dr. Majkrzak, We are pleased to inform you that your manuscript has been accepted for publication as a Regular Article in Physical Review A