Search

Numerous fairness metrics have been proposed and employed by artificial intelligence (AI) experts to quantitatively measure bias and define fairness in AI models. Recognizing the need to accommodate stakeholders' diverse fairness understandings, efforts are underway to solicit their input. However, conveying AI fairness metrics to stakeholders without AI expertise, capturing their personal preferences, and seeking a collective consensus remain challenging and underexplored. To bridge this gap, we propose a new framework, EARN Fairness, which facilitates collective metric decisions among stakeholders without requiring AI expertise. The framework features an adaptable interactive system and a stakeholder-centered EARN Fairness process to Explain fairness metrics, Ask stakeholders' personal metric preferences, Review metrics collectively, and Negotiate a consensus on metric selection. To gather empirical results, we applied the framework to a credit rating scenario and conducted a user study involving 18 decision subjects without AI knowledge. We identify their personal metric preferences and their acceptable level of unfairness in individual sessions. Subsequently, we uncovered how they reached metric consensus in team sessions. Our work shows that the EARN Fairness framework enables stakeholders to express personal preferences and reach consensus, providing practical guidance for implementing human-centered AI fairness in high-risk contexts. Through this approach, we aim to harmonize fairness expectations of diverse stakeholders, fostering more equitable and inclusive AI fairness.

Microvascular anatomy is known to be involved in various neurological disorders. However, understanding these disorders is hindered by the lack of imaging modalities capable of capturing the comprehensive three-dimensional vascular network structure at microscopic resolution. With a lateral resolution of $<=$20 {\textmu}m and ability to reconstruct large tissue blocks up to tens of cubic centimeters, serial-section optical coherence tomography (sOCT) is well suited for this task. This method uses intrinsic optical properties to visualize the vessels and therefore does not possess a specific contrast, which complicates the extraction of accurate vascular models. The performance of traditional vessel segmentation methods is heavily degraded in the presence of substantial noise and imaging artifacts and is sensitive to domain shifts, while convolutional neural networks (CNNs) require extensive labeled data and are also sensitive the precise intensity characteristics of the data that they are trained on. Building on the emerging field of synthesis-based training, this study demonstrates a synthesis engine for neurovascular segmentation in sOCT images. Characterized by minimal priors and high variance sampling, our highly generalizable method tested on five distinct sOCT acquisitions eliminates the need for manual annotations while attaining human-level precision. Our approach comprises two phases: label synthesis and label-to-image transformation. We demonstrate the efficacy of the former by comparing it to several more realistic sets of training labels, and the latter by an ablation study of synthetic noise and artifact models., Comment: 10 pages, 10 figures

We present the Keras domain packages KerasCV and KerasNLP, extensions of the Keras API for Computer Vision and Natural Language Processing workflows, capable of running on either JAX, TensorFlow, or PyTorch. These domain packages are designed to enable fast experimentation, with a focus on ease-of-use and performance. We adopt a modular, layered design: at the library's lowest level of abstraction, we provide building blocks for creating models and data preprocessing pipelines, and at the library's highest level of abstraction, we provide pretrained ``task" models for popular architectures such as Stable Diffusion, YOLOv8, GPT2, BERT, Mistral, CLIP, Gemma, T5, etc. Task models have built-in preprocessing, pretrained weights, and can be fine-tuned on raw inputs. To enable efficient training, we support XLA compilation for all models, and run all preprocessing via a compiled graph of TensorFlow operations using the tf.data API. The libraries are fully open-source (Apache 2.0 license) and available on GitHub., Comment: Submitted to Journal of Machine Learning Open Source Software

Serial sectioning Optical Coherence Tomography (sOCT) is a high-throughput, label free microscopic imaging technique that is becoming increasingly popular to study post-mortem neurovasculature. Quantitative analysis of the vasculature requires highly accurate segmentation; however, sOCT has low signal-to-noise-ratio and displays a wide range of contrasts and artifacts that depend on acquisition parameters. Furthermore, labeled data is scarce and extremely time consuming to generate. Here, we leverage synthetic datasets of vessels to train a deep learning segmentation model. We construct the vessels with semi-realistic splines that simulate the vascular geometry and compare our model with realistic vascular labels generated by constrained constructive optimization. Both approaches yield similar Dice scores, although with very different false positive and false negative rates. This method addresses the complexity inherent in OCT images and paves the way for more accurate and efficient analysis of neurovascular structures., Comment: 5 Pages, 2 figures. Accepted by Medical Imaging with Deep Learning

French and American participants listened to new music stimuli and evaluated the stimuli using either adjectives or quantitative musical dimensions. Results were analyzed using correspondence analysis (CA), hierarchical cluster analysis (HCA), multiple factor analysis (MFA), and partial least squares correlation (PLSC). French and American listeners differed when they described the musical stimuli using adjectives, but not when using the quantitative dimensions. The present work serves as a case study in research methodology that allows for a balance between relaxing experimental control and maintaining statistical rigor.

The kagome superconductor CsV$_3$Sb$_5$ hosts a variety of charge density wave (CDW) phases, which play a fundamental role in the formation of other exotic electronic instabilities. However, identifying the precise structure of these CDW phases and their intricate relationships remain the subject of intense debate, due to the lack of static probes that can distinguish the CDW phases with identical spatial periodicity. Here, we unveil the competition between two coexisting $2\times2\times2$ CDWs in CsV$_3$Sb$_5$ harnessing time-resolved X-ray diffraction. By analyzing the light-induced changes in the intensity of CDW superlattice peaks, we demonstrate the presence of both phases, each displaying a significantly different amount of melting upon excitation. The anomalous light-induced sharpening of peak width further shows that the phase that is more resistant to photo-excitation exhibits an increase in domain size at the expense of the other, thereby showcasing a hallmark of phase competition. Our results not only shed light on the interplay between the multiple CDW phases in CsV$_3$Sb$_5$, but also establish a non-equilibrium framework for comprehending complex phase relationships that are challenging to disentangle using static techniques., Comment: 10 pages, 4 figures with supplemental Material

We report on the evolution of the charge density wave (CDW) and spin density wave (SDW) order of a chromium film following photoexcitation with an ultrafast optical laser pulse. The CDW is measured by ultrafast time-resolved x-ray diffraction of the CDW satellite that tracks the suppression and recovery of the CDW following photoexcitation. We find that as the temperature of the film approaches a discontinuous phase transition in the CDW and SDW order, the time scales of recovery increase exponentially from the expected thermal time scales. We extend a Landau model for SDW systems to account for this critical slowing with the appropriate boundary conditions imposed by the geometry of the thin film system. This model allows us to assess the energy barrier between available CDW/SDW states with different spatial periodicities., Comment: Author typo fixed

A central paradigm of non-equilibrium physics concerns the dynamics of heterogeneity and disorder, impacting processes ranging from the behavior of glasses to the emergent functionality of active matter. Understanding these complex mesoscopic systems requires probing the microscopic trajectories associated with irreversible processes, the role of fluctuations and entropy growth, and the timescales on which non-equilibrium responses are ultimately maintained. Approaches that illuminate these processes in model systems may enable a more general understanding of other heterogeneous non-equilibrium phenomena, and potentially define ultimate speed and energy cost limits for information processing technologies. Here, we apply ultrafast single shot x-ray photon correlation spectroscopy to resolve the non-equilibrium, heterogeneous, and irreversible mesoscale dynamics during a light-induced phase transition. This approach defines a new way of capturing the nucleation of the induced phase, the formation of transient mesoscale defects at the boundaries of the nuclei, and the eventual annihilation of these defects, even in systems with complex polarization topologies. A non-equilibrium response spanning >10 orders of magnitude in timescales is observed, with multistep behavior similar to the plateaus observed in supercooled liquids and glasses. We show how the observed time-dependent long-time correlations can be understood in terms of the stochastic dynamics of domain walls, encoded in effective waiting-time distributions with power-law tails. This work defines new possibilities for probing the non-equilibrium and correlated dynamics of disordered and heterogeneous media.

In the age of personal voice assistants, the question of privacy arises. These digital companions often lack memory of past interactions, while relying heavily on the internet for speech processing, raising privacy concerns. Modern smartphones now enable on-device speech processing, making cloud-based solutions unnecessary. Personal assistants for the elderly should excel at memory recall, especially in medical examinations. The e-ViTA project developed a versatile conversational application with local processing and speaker recognition. This paper highlights the importance of speaker diarization enriched with sensor data fusion for contextualized conversation preservation. The use cases applied to the e-VITA project have shown that truly personalized dialogue is pivotal for individual voice assistants. Secure local processing and sensor data fusion ensure virtual companions meet individual user needs without compromising privacy or data security., Comment: 10 pages, 1 figure, to be presented at https://ihiet-ai.org/, Lausanne in April 2024

We demonstrate that the absorption of femtosecond x-ray pulses can excite quasi-spherical high-wavevector coherent acoustic phonon wavepackets using an all x-ray pump and probe scattering experiment. The time- and momentum-resolved diffuse scattering signal is consistent with strain pulses induced by the rapid electron cascade dynamics following photoionization at uncorrelated excitation centers. We quantify key parameters of this process, including the localization size of the strain wavepacket and the energy absorption efficiency, which are determined by the photoelectron and Auger electron cascade dynamics, as well as the electron-phonon interaction. In particular, we obtain the localization size of the observed strain wave packet to be 1.5 and 2.5 nm for bulk SrTiO$_3$ and KTaO$_3$ single crystals, even though there are no nanoscale structures or light-intensity patterns that would ordinarily be required to generate acoustic waves of wavelengths much shorter than the penetration depth. Whereas in GaAs and GaP we do not observe a signal above background. The results provide crucial information on x-ray matter interactions, which sheds light on the mechanism of x-ray energy deposition, and the study of high wavevector acoustic phonons and thermal transport at the nanoscale.

Ultrafast X-ray diffraction by the LCLS free-electron laser has been used to probe Charge Density Wave (CDW) systems under applied external currents. At sufficiently low currents, CDW wavefronts bend in the direction transverse to the 2k$_F$ wave vector. We show that this shear effect has the ability to focus or defocus hard X-ray beams, depending of the current direction, making it an electronic lens of a new kind, tunable at will from the Fraunhofer to the Fresnel regime. The effect is interpreted using the fractional Fourier transform showing how the macroscopic curvature of a nanometric modulation can be beneficially used to modify the propagation of X-ray beams., Comment: 3 pages, 3 figures

As the French, European and worldwide populations are aging, there is a strong interest for new systems that guarantee a reliable and privacy preserving home monitoring for frailty prevention. This work is a part of a global environmental audio analysis system which aims to help identification of Activities of Daily Life (ADL) through human and everyday life sounds recognition, speech presence and number of speakers detection. The focus is made on the number of speakers detection. In this article, we present how recent advances in sound processing and speaker diarization can improve the existing embedded systems. We study the performances of two new methods and discuss the benefits of DNN based approaches which improve performances by about 100%., Comment: JETSAN, Jun 2023, Aubervilliers & Paris, France

Quantifying electron-phonon interactions for the surface states of topological materials can provide key insights into surface-state transport, topological superconductivity, and potentially how to manipulate the surface state using a structural degree of freedom. We perform time-resolved x-ray diffraction (XRD) and angle-resolved photoemission (ARPES) measurements on Bi$_2$Te$_3$ and Bi$_2$Se$_3$, following the excitation of coherent A$_{1g}$ optical phonons. We extract and compare the deformation potentials coupling the surface electronic states to local A$_{1g}$-like displacements in these two materials using the experimentally determined atomic displacements from XRD and electron band shifts from ARPES.We find the coupling in Bi$_2$Te$_3$ and Bi$_2$Se$_3$ to be similar and in general in agreement with expectations from density functional theory. We establish a methodology that quantifies the mode-specific electron-phonon coupling experimentally, allowing detailed comparison to theory. Our results shed light on fundamental processes in topological insulators involving electron-phonon coupling.

Speaker diarization, or the task of finding "who spoke and when", is now used in almost every speech processing application. Nevertheless, its fairness has not yet been evaluated because there was no protocol to study its biases one by one. In this paper we propose a protocol and a scoring method designed to evaluate speaker diarization fairness. This protocol is applied on a large dataset of spoken utterances and report the performances of speaker diarization depending on the gender, the age, the accent of the speaker and the length of the spoken sentence. Some biases induced by the gender, or the accent of the speaker were identified when we applied a state-of-the-art speaker diarization method.

Interatomic forces that bind materials are largely determined by an often complex interplay between the electronic band-structure and the atomic arrangements to form its equilibrium structure and dynamics. As these forces also determine the phonon dispersion, lattice dynamics measurements are often crucial tools for understanding how materials transform between different structures. This is the case for the mono-chalcogenides which feature a number of lattice instabilities associated with their network of resonant bonds and a large tunability in their functional properties. SnSe hosts a novel lattice instability upon above-bandgap photoexcitation that is distinct from the distortions associated with its high temperature phase transition, demonstrating that photoexcitation can alter the interatomic forces significantly different than thermal excitation. Here we report decisive time-resolved X-ray scattering-based measurements of the nonequlibrium lattice dynamics in SnSe. By fitting interatomic force models to the excited-state dispersion, we determine this instability as being primarily due to changes in the fourth-nearest neighbor bonds that connect bilayers, with relatively little change to the intralayer resonant bonds. In addition to providing critical insight into the nonthermal bonding origin of the instability in SnSe, such measurements will be crucial for understanding and controlling materials properties under non-equilibrium conditions.

We investigate coupled electron-lattice dynamics in the topological insulator Bi2Te3 with time-resolved photoemission and time-resolved x-ray diffraction. It is well established that coherent phonons can be launched by optical excitation, but selection rules generally restrict these modes to zone-center wavevectors and Raman-active branches. We find that the topological surface state couples to additional modes, including a continuum of surface-projected bulk modes from both Raman- and infrared-branches, with possible contributions from surface-localized modes when they exist. Our calculations show that this surface vibrational spectrum occurs naturally as a consequence of the translational and inversion symmetries broken at the surface, without requiring the splitting-off of surface-localized phonon modes. The generality of this result suggests that coherent phonon spectra are useful by providing unique fingerprints for identifying surface states in more controversial materials. These effects may also expand the phase space for tailoring surface state wavefunctions via ultrafast optical excitation.

To evaluate electrostatics interactions, Molecular dynamics (MD) simulations rely on Particle Mesh Ewald (PME), an O(Nlog(N)) algorithm that uses Fast Fourier Transforms (FFTs) or, alternatively, on O(N) Fast Multipole Methods (FMM) approaches. However, the FFTs low scalability remains a strong bottleneck for large-scale PME simulations on supercomputers. On the opposite, - FFT-free - FMM techniques are able to deal efficiently with such systems but they fail to reach PME performances for small- to medium-size systems, limiting their real-life applicability. We propose ANKH, a strategy grounded on interpolated Ewald summations and designed to remain efficient/scalable for any size of systems. The method is generalized for distributed point multipoles and so for induced dipoles which makes it suitable for high performance simulations using new generation polarizable force fields towards exascale computing.

We report ultrafast x-ray scattering experiments of the quasi-1D charge density wave (CDW) material (TaSe$_4$)$_2$I following photoexcitation with femtosecond infrared laser pulses. From the time-dependent diffraction signal at the CDW sidebands we identify an amplitude mode derived primarily from the transverse acoustic component of the CDW static distortion. The dynamics of this acoustic amplitude mode are described well by a model of a displacive excitation, which we interpret as mediated through a coupling to the optical phonon component associated with the tetramerization of the Ta chains.

We introduce two new approximation methods for the numerical evaluation of the long-range Coulomb potential and the approximation of the resulting high dimensional Two-Electron Integrals tensor (TEI) with long-range interactions arising in molecular simulations. The first method exploits the tensorized structure of the compressed two-electron integrals obtained through two-dimensional Chebyshev interpolation combined with Gaussian quadrature. The second method is based on the Fast Multipole Method (FMM). Numerical experiments for different medium size molecules on high quality basis sets outline the efficiency of the two methods. Detailed algorithmic is provided in this paper as well as numerical comparison of the introduced approaches.

The advent of next-generation X-ray free electron lasers will be capable of delivering X-rays at a repetition rate approaching 1 MHz continuously. This will require the development of data systems to handle experiments at these type of facilities, especially for high throughput applications, such as femtosecond X-ray crystallography and X-ray photon fluctuation spectroscopy. Here, we demonstrate a framework which captures single shot X-ray data at the LCLS and implements a machine-learning algorithm to automatically extract the contrast parameter from the collected data. We measure the time required to return the results and assess the feasibility of using this framework at high data volume. We use this experiment to determine the feasibility of solutions for `live' data analysis at the MHz repetition rate.

The structures, strain fields, and defect distributions in solid materials underlie the mechanical and physical properties across numerous applications. Many modern microstructural microscopy tools characterize crystal grains, domains and defects required to map lattice distortions or deformation, but are limited to studies of the (near) surface. Generally speaking, such tools cannot probe the structural dynamics in a way that is representative of bulk behavior. Synchrotron X-ray diffraction based imaging has long mapped the deeply embedded structural elements, and with enhanced resolution, Dark Field X-ray Microscopy (DFXM) can now map those features with the requisite nm-resolution. However, these techniques still suffer from the required integration times due to limitations from the source and optics. This work extends DFXM to X-ray free electron lasers, showing how the $10^{12}$ photons per pulse available at these sources offer structural characterization down to 100 fs resolution (orders of magnitude faster than current synchrotron images). We introduce the XFEL DFXM setup with simultaneous bright field microscopy to probe density changes within the same volume. This work presents a comprehensive guide to the multi-modal ultrafast high-resolution X-ray microscope that we constructed and tested at two XFELs, and shows initial data demonstrating two timing strategies to study associated reversible or irreversible lattice dynamics.

Evaluating the COCO mean average precision (MaP) and COCO recall metrics as part of the static computation graph of modern deep learning frameworks poses a unique set of challenges. These challenges include the need for maintaining a dynamic-sized state to compute mean average precision, reliance on global dataset-level statistics to compute the metrics, and managing differing numbers of bounding boxes between images in a batch. As a consequence, it is common practice for researchers and practitioners to evaluate COCO metrics as a post training evaluation step. With a graph-friendly algorithm to compute COCO Mean Average Precision and recall, these metrics could be evaluated at training time, improving visibility into the evolution of the metrics through training curve plots, and decreasing iteration time when prototyping new model versions. Our contributions include an accurate approximation algorithm for Mean Average Precision, an open source implementation of both COCO mean average precision and COCO recall, extensive numerical benchmarks to verify the accuracy of our implementations, and an open-source training loop that include train-time evaluation of mean average precision and recall., Comment: 7 pages, 3 figures

Language models demonstrate both quantitative improvement and new qualitative capabilities with increasing scale. Despite their potentially transformative impact, these new capabilities are as yet poorly characterized. In order to inform future research, prepare for disruptive new model capabilities, and ameliorate socially harmful effects, it is vital that we understand the present and near-future capabilities and limitations of language models. To address this challenge, we introduce the Beyond the Imitation Game benchmark (BIG-bench). BIG-bench currently consists of 204 tasks, contributed by 450 authors across 132 institutions. Task topics are diverse, drawing problems from linguistics, childhood development, math, common-sense reasoning, biology, physics, social bias, software development, and beyond. BIG-bench focuses on tasks that are believed to be beyond the capabilities of current language models. We evaluate the behavior of OpenAI's GPT models, Google-internal dense transformer architectures, and Switch-style sparse transformers on BIG-bench, across model sizes spanning millions to hundreds of billions of parameters. In addition, a team of human expert raters performed all tasks in order to provide a strong baseline. Findings include: model performance and calibration both improve with scale, but are poor in absolute terms (and when compared with rater performance); performance is remarkably similar across model classes, though with benefits from sparsity; tasks that improve gradually and predictably commonly involve a large knowledge or memorization component, whereas tasks that exhibit "breakthrough" behavior at a critical scale often involve multiple steps or components, or brittle metrics; social bias typically increases with scale in settings with ambiguous context, but this can be improved with prompting., Comment: 27 pages, 17 figures + references and appendices, repo: https://github.com/google/BIG-bench