Search

In this paper, we introduce and study McKean-Vlasov processes of bridge type. Specifically, we examine a stochastic differential equation (SDE) of the form: $$\mathrm{d} \xi_t=-\mu(t,\mathbb{E}[\varphi_1(\xi_t)]) \frac{\xi_t}{T-t} \mathrm{d} t+\sigma(t,\mathbb{E}[\varphi_2(\xi_t)]) \mathrm{d} W_t,\,\, t

In-cache computing technology transforms existing caches into long-vector compute units and offers low-cost alternatives to building expensive vector engines for mobile CPUs. Unfortunately, existing long-vector Instruction Set Architecture (ISA) extensions, such as RISC-V Vector Extension (RVV) and Arm Scalable Vector Extension (SVE), provide only one-dimensional strided and random memory accesses. While this is sufficient for typical vector engines, it fails to effectively utilize the large Single Instruction, Multiple Data (SIMD) widths of in-cache vector engines. This is because mobile data-parallel kernels expose limited parallelism across a single dimension. Based on our analysis of mobile vector kernels, we introduce a long-vector Multi-dimensional Vector ISA Extension (MVE) for mobile in-cache computing. MVE achieves high SIMD resource utilization and enables flexible programming by abstracting cache geometry and data layout. The proposed ISA features multi-dimensional strided and random memory accesses and efficient dimension-level masked execution to encode parallelism across multiple dimensions. Using a wide range of data-parallel mobile workloads, we demonstrate that MVE offers significant performance and energy reduction benefits of 2.9x and 8.8x, on average, compared to the SIMD units of a commercial mobile processor, at an area overhead of 3.6%., Comment: 2025 IEEE International Symposium on High-Performance Computer Architecture (HPCA)

Recently, there has been growing concern about heavy-tailed and skewed noise in biological data. We introduce RobustPALMRT, a flexible permutation framework for testing the association of a covariate of interest adjusted for control covariates. RobustPALMRT controls type I error rate for finite-samples, even in the presence of heavy-tailed or skewed noise. The new framework expands the scope of state-of-the-art tests in three directions. First, our method applies to robust and quantile regressions, even with the necessary hyper-parameter tuning. Second, by separating model-fitting and model-evaluation, we discover that performance improves when using a robust loss function in the model-evaluation step, regardless of how the model is fit. Third, we allow fitting multiple models to detect specialized features of interest in a distribution. To demonstrate this, we introduce DispersionPALRMT, which tests for differences in dispersion between treatment and control groups. We establish theoretical guarantees, identify settings where our method has greater power than existing methods, and analyze existing immunological data on Long-COVID patients. Using RobustPALMRT, we unveil novel differences between Long-COVID patients and others even in the presence of highly skewed noise.

Large Language Model (LLM)-Powered Conversational Agents have the potential to provide users with scaled behavioral healthcare support, and potentially even deliver full-scale "AI therapy'" in the future. While such agents can already conduct fluent and proactive emotional support conversations, they inherently lack the ability to (a) consistently and reliably act by predefined rules to align their conversation with an overarching therapeutic concept and (b) make their decision paths inspectable for risk management and clinical evaluation -- both essential requirements for an "AI Therapist". In this work, we introduce a novel paradigm for dialog policy planning in conversational agents enabling them to (a) act according to an expert-written "script" that outlines the therapeutic approach and (b) explicitly transition through a finite set of states over the course of the conversation. The script acts as a deterministic component, constraining the LLM's behavior in desirable ways and establishing a basic architecture for an AI Therapist. We implement two variants of Script-Based Dialog Policy Planning using different prompting techniques and synthesize a total of 100 conversations with LLM-simulated patients. The results demonstrate the feasibility of this new technology and provide insights into the efficiency and effectiveness of different implementation variants., Comment: 9 pages, 5 figures, 1 table

The joint implementation of Federated learning (FL) and Explainable artificial intelligence (XAI) will allow training models from distributed data and explaining their inner workings while preserving important aspects of privacy. Towards establishing the benefits and tensions associated with their interplay, this scoping review maps those publications that jointly deal with FL and XAI, focusing on publications where an interplay between FL and model interpretability or post-hoc explanations was found. In total, 37 studies met our criteria, with more papers focusing on explanation methods (mainly feature relevance) than on interpretability (mainly algorithmic transparency). Most works used simulated horizontal FL setups involving 10 or fewer data centers. Only one study explicitly and quantitatively analyzed the influence of FL on model explanations, revealing a significant research gap. Aggregation of interpretability metrics across FL nodes created generalized global insights at the expense of node-specific patterns being diluted. 8 papers addressed the benefits of incorporating explanation methods as a component of the FL algorithm. Studies using established FL libraries or following reporting guidelines are a minority. More quantitative research and structured, transparent practices are needed to fully understand their mutual impact and under which conditions it happens., Comment: 16 pages, 11 figures, submitted in IEEE Trans. Knowledge and Data Engineering

Magnon-condensation, which emerges in pumped bosonic systems at room temperature, continues to garner great interest for its long-lived coherence. While traditionally formulated in terms of Bose-Einstein condensation, which typically occurs at ultra-low temperatures, it could potentially also be explained by Fr\"ohlich-condensation, a hypothesis of Bose-Einstein-like condensation in living systems at ambient temperatures. This prompts general questions relating to fundamental differences between coherence phenomena in open and isolated quantum systems. To that end, we introduce a simple model of bosonic condensation in an open quantum system (OQS) formulation, wherein bosons dissipatively interact with an oscillator (phonon) bath. Our derived equations of motion for expected boson occupations turns out to be similar in form to the rate equations governing Fr\"ohlich-condensation. Provided that specific system parameters result in correlations that amplify or diminish the condensation effects, we thereby posit that our treatment offers a better description of high-temperature condensation compared to traditional formulations obtained using equilibrium thermodynamics. By comparing our OQS derivation with the original uncorrelated and previous semi-classical rate equations, we furthermore highlight how both classical anti-correlations and quantum correlations alter the bosonic occupation distribution., Comment: 7 pages, 2 figures, plus supplementary material (8 pages)

We address the generic problem of optimal quantum state preparation for open quantum systems. It is well known that open quantum systems can be simulated by quantum trajectories described by a stochastic Schr\"odinger equation. In this context, the state preparation becomes a stochastic optimal control (SOC) problem. The latter requires the solution of the Hamilton-Jacobi-Bellman equation, which is, in general, challenging to solve. A notable exception are the so-called path integral (PI) control problems, for which one can estimate the optimal control solution by direct sampling of the cost objective. In this work, we derive a class of quantum state preparation problems that are amenable to PI control techniques, and propose a corresponding algorithm, which we call Quantum Diffusion Control (QDC). Unlike conventional quantum control algorithms, QDC avoids computing gradients of the cost function to determine the optimal control. Instead, it employs adaptive importance sampling, a technique where the controls are iteratively improved based on global averages over quantum trajectories. We also demonstrate that QDC, used as an annealer in the environmental coupling strength, finds high accuracy solutions for unitary (noiseless) quantum control problems. We further discuss the implementation of this technique on quantum hardware. We illustrate the effectiveness of our approach through examples of open-loop control for single- and multi-qubit systems., Comment: We added a new section on annealing control, supplementary material, and formatted for journal submission

Cerebrovascular disease often requires multiple imaging modalities for accurate diagnosis, treatment, and monitoring. Computed Tomography Angiography (CTA) and Time-of-Flight Magnetic Resonance Angiography (TOF-MRA) are two common non-invasive angiography techniques, each with distinct strengths in accessibility, safety, and diagnostic accuracy. While CTA is more widely used in acute stroke due to its faster acquisition times and higher diagnostic accuracy, TOF-MRA is preferred for its safety, as it avoids radiation exposure and contrast agent-related health risks. Despite the predominant role of CTA in clinical workflows, there is a scarcity of open-source CTA data, limiting the research and development of AI models for tasks such as large vessel occlusion detection and aneurysm segmentation. This study explores diffusion-based image-to-image translation models to generate synthetic CTA images from TOF-MRA input. We demonstrate the modality conversion from TOF-MRA to CTA and show that diffusion models outperform a traditional U-Net-based approach. Our work compares different state-of-the-art diffusion architectures and samplers, offering recommendations for optimal model performance in this cross-modality translation task.

The competition between the Kondo spin-screening effect and the Ruderman-Kittel-Kasuya-Yosida (RKKY) interaction in heavy-fermion systems drives the quantum phase transition between the magnetically ordered and the heavy-Fermi-liquid ground states. Despite intensive investigations of heavy quasiparticles on the Kondo-screened side of the quantum phase transition and of their breakdown at the quantum critical point, the magnetically ordering side has not systematically been studied. Using terahertz time-domain spectroscopy, we report a suppression of the Kondo quasiparticle weight in CeCu$_{6-x}$Au$_x$ samples on the antiferromagnetic side of the quantum phase transition at temperatures as much as two orders of magnitude above the N\'{e}el temperature $T_\text{N}$. With our systematic investigations into the high-temperature, paramagnetic region on the antiferromagnetic side of the phase diagram of CeCu$_{6-x}$Au$_x$, i.e., with $x =$ 0.2, 0.3, and 0.5, we show that the suppression results from a quantum frustration effect induced by the temperature-independent RKKY interaction. Hence, our results emphasize that besides critical fluctuations, the RKKY interaction may play an important role in the quantum-critical scenario., Comment: 6 pages, 4 figures

Bots have become increasingly prevalent in the digital sphere and have taken up a proactive role in shaping democratic processes. While previous studies have focused on their influence at the individual level, their potential macro-level impact on communication dynamics is still little understood. This study adopts an information theoretic approach from dynamical systems theory to examine the role of political bots shaping the dynamics of an online political discussion on Twitter. We quantify the components of this dynamic process in terms of its complexity, predictability, and the remaining uncertainty. Our findings suggest that bot activity is associated with increased complexity and uncertainty in the structural dynamics of online political communication. This work serves as a showcase for the use of information-theoretic measures from dynamical systems theory in modeling human-bot dynamics as a computational process that unfolds over time.

Life finds a way. For sessile organisms like plants, the need to adapt to changes in the environment is even more poignant. For humanity, the need to develop crops that can grow in diverse environments and feed our growing population is an existential one. The advent of the genomics era enabled the generation of high-throughput data and computational methods that serve as powerful hypothesis-generating tools to understand the genomic and gene functional basis of stress resilience. Today, the proliferation of artificial intelligence (AI) allows scientists to rapidly screen through high-throughput datasets to uncover elusive patterns and correlations, enabling us to create more performant models for prediction and hypothesis generation in plant biology. This review aims to provide an overview of the availability of large-scale data in plant stress research and discuss the application of AI tools on these large-scale datasets in a bid to develop more stress-resilient plants.

The thalamus is a subcortical structure of central importance to brain function, which is organized in smaller nuclei with specialized roles. Despite significant functional and clinical relevance, locating and distinguishing the different thalamic nuclei in vivo, non-invasively, has proved challenging with conventional imaging techniques, such as T$_{1}$ and T$_{2}$-weighted magnetic resonance imaging (MRI). This key limitation has prompted extensive research efforts, and several new candidate MRI sequences for thalamic imaging have been proposed, especially at 7T. However, studies to date have mainly been centered on individual techniques, and often focused on subsets of specific nuclei. It is now critical to evaluate which options are best for which nuclei, and which are globally the most informative. This work addresses these questions through a comprehensive evaluation of thalamic structural imaging techniques in humans at 7T, including several variants of T$_{1}$, T$_{2}$, T$_{2}$* and magnetic susceptibility-based contrasts. All images were obtained from the same participants, to allow direct comparisons without anatomical variability confounds. The different contrasts were qualitatively and quantitatively analyzed with dedicated approaches, referenced to well-established thalamic atlases. Overall, the analyses showed that quantitative susceptibility mapping (QSM) and T$_{1}$-weighted MP2RAGE tuned to maximize gray-to-white matter contrast are currently the most valuable options. The two contrasts display unique, complementary features and, together, enable the distinction of the majority of known nuclei. Likewise, their combined information could provide a powerful input for automatic segmentation approaches. To our knowledge, this study represents the most comprehensive assessment of structural MRI contrasts for thalamic imaging to date., Comment: 39 pages, 6 figures, 3 tables, 8 supplementary figures

Large Language Models (LLMs), such as ChatGPT, have taken the world by storm and have passed certain forms of the Turing test. However, LLMs are not limited to human language and analyze sequential data, such as DNA, protein, and gene expression. The resulting foundation models can be repurposed to identify the complex patterns within the data, resulting in powerful, multi-purpose prediction tools able to explain cellular systems. This review outlines the different types of LLMs and showcases their recent uses in biology. Since LLMs have not yet been embraced by the plant community, we also cover how these models can be deployed for the plant kingdom.

In strongly correlated systems such as heavy-fermion materials, the coherent superposition of localized and mobile spin states leads to the formation of Kondo resonant states, which on a dense, periodic array of Kondo ions develop lattice coherence. Characteristically, these quantum-coherent superposition states respond to a terahertz (THz) excitation by a delayed THz pulse on the scale of the material's Kondo energy scale and, hence, independent of the pump-light intensity. However, delayed response is also typical for superradiance in an ensemble of excited atoms. In this case, quantum coherence is established by the coupling to an external, electromagnetic mode and, hence, dependent on the pump-light intensity. In the present work, we investigate the physical origin of the delayed pulse, i.e., inherent, correlation-induced versus light-induced coherence, in the prototypical heavy-fermion compound CeCu_5.9Au_0.1. We study the delay, duration and amplitude of the THz pulse at various temperatures in dependence on the electric-field strength of the incident THz excitation, ranging from 0.3 to 15.2 kV/cm. We observe a robust delayed response at approximately 6 ps with an amplitude proportional to the amplitude of the incident THz wave. This is consistent with theoretical expectation for the Kondo-like coherence and thus provides compelling evidence for the dominance of condensed-matter versus optical coherence in the heavy-fermion compound., Comment: 6 pages, 2 figures

In the biomedical environment, experiments assessing dynamic processes are primarily performed by a human acquisition supervisor. Contemporary implementations of such experiments frequently aim to acquire a maximum number of relevant events from sometimes several hundred parallel, non-synchronous processes. Since in some high-throughput experiments, only one or a few instances of a given process can be observed simultaneously, a strategy for planning and executing an efficient acquisition paradigm is essential. To address this problem, we present two new methods in this paper. The first method, Encoded Dynamic Process (EDP), is Artificial Intelligence (AI)-based and represents dynamic processes so as to allow prediction of pseudo-time values from single still images. Second, with Experiment Automation Pipeline for Dynamic Processes (EAPDP), we present a Machine Learning Operations (MLOps)-based pipeline that uses the extracted knowledge from EDP to efficiently schedule acquisition in biomedical experiments for dynamic processes in practice. In a first experiment, we show that the pre-trained State-Of-The- Art (SOTA) object segmentation method Contour Proposal Networks (CPN) works reliably as a module of EAPDP to extract the relevant object for EDP from the acquired three-dimensional image stack., Comment: Proceedings - 33. Workshop Computational Intelligence

Stochastically fluctuating multi-well systems as physical implementations of energy-based machine learning models promise a route towards neuromorphic hardware. Understanding the response of multi-well systems to dynamic input signals is crucial in this regard. Here, we investigate the stochastic response of binary orbital memory states derived from individual Fe and Co atoms on a black phosphorus surface to sinusoidal input voltages. Using scanning tunneling microscopy, we quantify the state residence times for DC and AC voltage drive with various input frequencies. We find that Fe and Co atoms both exhibit features of synchronization to the AC input, but only Fe atoms demonstrate a significant frequency-dependent change in the time-averaged state occupations. By modeling the underlying stochastic process, we show that the frequency response of the system is directly related to the DC voltage dependence of the state asymmetry. This relation provides a tunable way to induce population changes in stochastic systems and lays the foundation for understanding the response of multi-well systems to dynamical input signals.

While supermassive black holes are ubiquitous features of galactic nuclei, only a small minority are observed during episodes of luminous accretion. The physical mechanism(s) driving the onset of fueling and ignition in these active galactic nuclei (AGN) are still largely unknown for many galaxies and AGN-selection criteria. Attention has focused on AGN triggering by means of major galaxy mergers gravitationally funneling gas towards the galactic center, with evidence both for and against this scenario. However, several recent studies have found that radio-loud AGN overwhelmingly reside in ongoing or recent major galaxy mergers. In this study, we test the hypothesis that major galaxy mergers are important triggers for radio-loud AGN activity in powerful quasars during cosmic noon (1 < z < 2). To this end, we compare Hubble Space Telescope WFC3/IR observations of the z > 1 3CR radio-loud broad-lined quasars to three matched radio-quiet quasar control samples. We find strong evidence for major-merger activity in nearly all radio-loud AGN, in contrast to the much lower merger fraction in the radio-quiet AGN. These results suggest major galaxy mergers are key ingredients to launching powerful radio jets. Given many of our radio-loud quasars are blue, our results present a possible challenge to the "blow-out" paradigm of galaxy evolution models in which blue quasars are the quiescent end result following a period of red quasar feedback initiated by a galaxy merger. Finally, we find a tight correlation between black hole mass and host galaxy luminosity for these different high-redshift AGN samples inconsistent with those observed for local elliptical galaxies., Comment: Published by ApJ

The quantum Boltzmann machine (QBM) is a generative machine learning model for both classical data and quantum states. Training the QBM consists of minimizing the relative entropy from the model to the target state. This requires QBM expectation values which are computationally intractable for large models in general. It is therefore important to develop heuristic training methods that work well in practice. In this work, we study a heuristic method characterized by a nested loop: the inner loop trains the $\beta$-variational quantum eigensolver ($\beta$-VQE) by Liu et al (2021 Mach. Learn.: Sci. Technol.2 025011) to approximate the QBM expectation values; the outer loop trains the QBM to minimize the relative entropy to the target. We show that low-rank representations obtained by $\beta$-VQE provide an efficient way to learn low-rank target states, such as classical data and low-temperature quantum tomography. We test the method on both classical and quantum target data with numerical simulations of up to 10 qubits. For the cases considered here, the obtained QBMs can model the target to high fidelity. We implement a trained model on a physical quantum device. The approach offers a valuable route towards variationally training QBMs on near-term quantum devices., Comment: 9 pages, 11 figures

Twenty-six years ago a small committee report, building on earlier studies, expounded a compelling and poetic vision for the future of astronomy, calling for an infrared-optimized space telescope with an aperture of at least $4m$. With the support of their governments in the US, Europe, and Canada, 20,000 people realized that vision as the $6.5m$ James Webb Space Telescope. A generation of astronomers will celebrate their accomplishments for the life of the mission, potentially as long as 20 years, and beyond. This report and the scientific discoveries that follow are extended thank-you notes to the 20,000 team members. The telescope is working perfectly, with much better image quality than expected. In this and accompanying papers, we give a brief history, describe the observatory, outline its objectives and current observing program, and discuss the inventions and people who made it possible. We cite detailed reports on the design and the measured performance on orbit., Comment: Accepted by PASP for the special issue on The James Webb Space Telescope Overview, 29 pages, 4 figures

We consider the optimal stopping time problem under model uncertainty $R(v)= {\text{ess}\sup\limits}_{ \mathbb{P} \in \mathcal{P}} {\text{ess}\sup\limits}_{\tau \in \mathcal{S}_v} E^\mathbb{P}[Y(\tau) \vert \mathcal{F}_v]$, for every stopping time $v$, set in the framework of families of random variables indexed by stopping times. This setting is more general than the classical setup of stochastic processes, and particularly allows for general payoff processes that are not necessarily right-continuous. Under weaker integrability, and regularity assumptions on the reward family $Y=(Y(v), v\in \mathcal{S})$, we show the existence of an optimal stopping time. We then proceed to find sufficient conditions for the existence of an optimal model. For this purpose, we present a universal Doob-Meyer-Mertens's decomposition for the Snell envelope family associated with $Y$ in the sense that it holds simultaneously for all $\mathbb{P} \in \mathcal{P}$. This decomposition is then employed to prove the existence of an optimal probability model and study its properties.