Search

This paper examines the impact, characteristics, and entrepreneurial proclivities of foreign-born college graduates in the United States. A significant body of research has found that immigrants are more likely to start businesses than those born in the U.S., and the propensity of international students to concentrate in STEM fields indicates enormous potential for economic contributions and innovation. Yet the static nature of the immigration system, with visa pathways and restrictions that discourage business creation, hamper the nation's ability to take full advantage of the benefits immigrants can provide. In fact, this study finds that the U.S. immigration system likely delays foreign-born graduates from creating incorporated firms by as many as five years. The authors estimate that the creation of 150,000 incorporated firms and 580,000 jobs were delayed between 2013 and 2021. Without reform, the U.S. will continue to depress high-value firm creation by international students and cease to be the primary destination of global talent.

Research and development work in artificial intelligence in education (AIED) is wide ranging and rapidly growing to support all areas of science, technology, engineering, and mathematics (STEM) teaching and learning. At the risk of hyperbole, this is potentially the most fundamentally game-changing technology for education to emerge since the internet. Building from decades of work on AI and AI-based learning and teaching technologies, the recent advances in AIED are pushing us to reimagine what is possible for STEM teaching and learning. AIED research initiatives are being speedily funded, and AIED advances are quickly becoming integrated into STEM education. It is transforming how teachers teach and how students learn. It is also transforming how education developers and researchers conduct their expansive work. There is excitement about the promise of AIED as well as growing concern that the breakthroughs in AIED are impacting everyday education practice in ways that may perpetuate long-standing biases and diminish the potential for positive outcomes. This brief is the first in a three-part series on AIED related to STEM research, teaching, and learning. The topics address ethical approaches to AI in STEM education research, AI for STEM teaching, and AI for STEM learning. This series is sponsored by the Community for Advancing Discovery Research in Education (CADRE), a National Science Foundation-funded network for STEM education researchers endeavoring to improve STEM teaching and learning through research, development, and various information-sharing and community-building mechanisms. Researchers in the CADRE network are part of a portfolio of projects funded through NSF's Discovery Research PreK-12 (DRK-12) program. The DRK-12 portfolio is wide-ranging, with a multitude of projects that focus on applied research and development to generate innovative research-informed and field-tested tools, products, and approaches that are intended to enhance STEM teaching and learning. Over the past several years, the portfolio has grown to include an increasing number of projects that leverage AIED to achieve their goals related to teaching or learning. It is expected to continue to grow. This series has been inspired by the question, "What are the essential considerations for researchers and developers who are designing, studying, and using AI in K-12 STEM?" Our hope is that the opportunities and challenges discussed in this series will generate reflection and rich discussion for the better and support the transformative use of AI to achieve positive and wide-reaching impact for all learners. In this first brief, "Toward Ethical and Just AI in Education Research," the authors are concerned with the ethical reasoning and decisions made in the development, study, and use of AIED technologies. Recognizing that AIED technologies reflect both the intended and unintended biases of the designers and the wider society, they advocate for the adoption of policies and practices that prioritize ethics, equity, and justice in research and development initiatives using AIED technologies in K-12 education. In an effort to provide guidance to researchers and developers, they lay the groundwork for responsible AI research and its implementation in educational settings. This foundation draws in part from the ethics rules for research with human subjects that have guided researchers for decades, but goes beyond this to frame a more all-encompassing stance rooted in justice and equity. The authors illustrate how ethical AI research can be strengthened by building from well-established ethical principles used in research and society at large. Taking into account these principles, they propose an ethical AIED framework and a set of tools that they have found to be supportive of continuous reflection, communication, and improvement toward inclusive and equitable AIED research and development. Their guidance is in the service of ensuring that the good intentions of researchers and developers will lead to positive design decisions and actions that create inclusive AIED technology products and systems. This is a valuable contribution that encourages a shift in focus to bring ethics, justice, and the values of communities of teachers, students, and families to the forefront of research and development practice.

Biologists study Diatoms, a fundamental algae, to assess the health of aquatic systems. Diatom specimens have traditionally been preserved on analog slides, where a single slide can contain thousands of these microscopic organisms. Digitization of these collections presents both metadata challenges and opportunities. This paper reports on metadata research aimed at providing access to a digital portion of the Academy of Natural Sciences' Diatom Herbarium, Drexel University. We report results of a 3-part study covering 1) a review of relevant metadata standards and a microscopy metadata framework shared by Hammer et al., 2) a baseline metadata alignment mapping current diatom metadata properties to standard metadata types, and 3) a metadata risk analysis associated with the course of standard data curation practices. This research is part of an effort involving the transfer of these digital slides to an new system, DataFed, to support global accessible. The final section of this paper includes a conclusion and discusses next steps., Comment: 13 pages, 2 figures, 1 table, to be published in MTSR 2024 conference proceedings

The first very long baseline interferometry (VLBI) detections at 870$\mu$m wavelength (345$\,$GHz frequency) are reported, achieving the highest diffraction-limited angular resolution yet obtained from the surface of the Earth, and the highest-frequency example of the VLBI technique to date. These include strong detections for multiple sources observed on inter-continental baselines between telescopes in Chile, Hawaii, and Spain, obtained during observations in October 2018. The longest-baseline detections approach 11$\,$G$\lambda$ corresponding to an angular resolution, or fringe spacing, of 19$\mu$as. The Allan deviation of the visibility phase at 870$\mu$m is comparable to that at 1.3$\,$mm on the relevant integration time scales between 2 and 100$\,$s. The detections confirm that the sensitivity and signal chain stability of stations in the Event Horizon Telescope (EHT) array are suitable for VLBI observations at 870$\mu$m. Operation at this short wavelength, combined with anticipated enhancements of the EHT, will lead to a unique high angular resolution instrument for black hole studies, capable of resolving the event horizons of supermassive black holes in both space and time., Comment: Corresponding author: S. Doeleman

Cosmic Explorer (CE) is a next-generation ground-based gravitational-wave observatory that is being designed in the 2020s and is envisioned to begin operations in the 2030s together with the Einstein Telescope in Europe. The CE concept currently consists of two widely separated L-shaped observatories in the United States, one with 40 km-long arms and the other with 20 km-long arms. This order of magnitude increase in scale with respect to the LIGO-Virgo-KAGRA observatories will, together with technological improvements, deliver an order of magnitude greater astronomical reach, allowing access to gravitational waves from remnants of the first stars and opening a wide discovery aperture to the novel and unknown. In addition to pushing the reach of gravitational-wave astronomy, CE endeavors to approach the lifecycle of large scientific facilities in a way that prioritizes mutually beneficial relationships with local and Indigenous communities. This article describes the (scientific, cost and access, and social) criteria that will be used to identify and evaluate locations that could potentially host the CE observatories., Comment: 12 pages, 1 figure

It has been recently shown that the celebrated SCFT$_4$/VOA$_2$ correspondence can be bridged via three-dimensional field theories arising from a specific R-symmetry twisted circle reduction. We apply this twisted reduction to the $(A_1,A_{n})$ and $(A_1,D_{n})$ families of 4d $\mathcal{N}=2$ Argyres-Douglas SCFTs using their $\mathcal{N}=1$ Agarwal-Maruyoshi-Song Lagrangians. From $(A_1,A_{2n})$ we derive the Gang-Kim-Stubbs family of 3d $\mathcal{N}=2$ gauge theories with SUSY enhancement to $\mathcal{N}=4$ in the infrared, generalizing a recent derivation made in the special cases $n=1,2$. Topological twists of these theories are known to yield $semisimple$ TQFTs supporting $rational$ VOAs on holomorphic boundaries. From $(A_1,A_{2n-1})$, $(A_1,D_{2n+1})$, and $(A_1,D_{2n})$, we obtain three new infinite families of 3d $\mathcal{N}=2$ abelian gauge theories, all with monopole superpotentials, flowing to $\mathcal{N}=4$ SCFTs without Coulomb branch, but with the same non-trivial Higgs branch as the four-dimensional parent. Their topological A-twist yields $non$-$semisimple$ TQFTs related to $logarithmic$ VOAs such as $\widehat{\mathfrak{su}}(2)_{-4/3}$., Comment: 46 pages + an appendix, 2 figures

In June 2022, the Gaia mission released a catalog of astrometric orbital solutions for 168,065 binary systems, by far the largest such catalog to date. Unlike previous binary stars catalogs, which were heterogeneous collections of orbits from different surveys and instruments, these orbits were derived using Gaia data alone. Despite this homogeneity, the selection function is difficult to characterize because of choices made in the construction of the catalog. Understanding the catalog's selection function is required to model and interpret its contents. We use a combination of analytic and empirical prescriptions to construct a function that computes the probability that a binary with a given set of properties would have been published in the Gaia Data Release 3 astrometric orbit catalog. We also construct a binary population synthesis model based on Moe & Di Stefano (2017) to validate our characterization of the selection function, finding good agreement with the actual Gaia NSS catalog, with the exception of the orbital eccentricity distribution. The NSS catalog suggests high-eccentricity orbits are relatively uncommon at intermediate periods $100 \lesssim P_{orb} \lesssim 1000$ days. As an example application of the selection function, we estimate the Gaia DR3 detection probabilities of the star + BH binaries Gaia BH1, BH2, and BH3. We also estimate the population of Sun-like star + BH binaries in the Galaxy to be $\sim 5000$ for $100 < P_{orb} < 400$ day, $\lesssim 2,000$ for $400 < P_{orb} < 1000$ day, and $ \lesssim 20,000$ for $1000 < P_{orb} < 2000$ days., Comment: 24 pages, 14 figures, 2 appendices. Submitted to ApJ, comments welcome. Selection function code will be available here: https://github.com/caseylam/dr3_nss_ast_orbits_selection

In real-world applications, the distribution of the data, and our goals, evolve over time. The prevailing theoretical framework for studying machine learning, namely probably approximately correct (PAC) learning, largely ignores time. As a consequence, existing strategies to address the dynamic nature of data and goals exhibit poor real-world performance. This paper develops a theoretical framework called "Prospective Learning" that is tailored for situations when the optimal hypothesis changes over time. In PAC learning, empirical risk minimization (ERM) is known to be consistent. We develop a learner called Prospective ERM, which returns a sequence of predictors that make predictions on future data. We prove that the risk of prospective ERM converges to the Bayes risk under certain assumptions on the stochastic process generating the data. Prospective ERM, roughly speaking, incorporates time as an input in addition to the data. We show that standard ERM as done in PAC learning, without incorporating time, can result in failure to learn when distributions are dynamic. Numerical experiments illustrate that prospective ERM can learn synthetic and visual recognition problems constructed from MNIST and CIFAR-10., Comment: Accepted to NeurIPS 2024

Technologies for recognizing facial attributes like race, gender, age, and emotion have several applications, such as surveillance, advertising content, sentiment analysis, and the study of demographic trends and social behaviors. Analyzing demographic characteristics based on images and analyzing facial expressions have several challenges due to the complexity of humans' facial attributes. Traditional approaches have employed CNNs and various other deep learning techniques, trained on extensive collections of labeled images. While these methods demonstrated effective performance, there remains potential for further enhancements. In this paper, we propose to utilize vision language models (VLMs) such as generative pre-trained transformer (GPT), GEMINI, large language and vision assistant (LLAVA), PaliGemma, and Microsoft Florence2 to recognize facial attributes such as race, gender, age, and emotion from images with human faces. Various datasets like FairFace, AffectNet, and UTKFace have been utilized to evaluate the solutions. The results show that VLMs are competitive if not superior to traditional techniques. Additionally, we propose "FaceScanPaliGemma"--a fine-tuned PaliGemma model--for race, gender, age, and emotion recognition. The results show an accuracy of 81.1%, 95.8%, 80%, and 59.4% for race, gender, age group, and emotion classification, respectively, outperforming pre-trained version of PaliGemma, other VLMs, and SotA methods. Finally, we propose "FaceScanGPT", which is a GPT-4o model to recognize the above attributes when several individuals are present in the image using a prompt engineered for a person with specific facial and/or physical attributes. The results underscore the superior multitasking capability of FaceScanGPT to detect the individual's attributes like hair cut, clothing color, postures, etc., using only a prompt to drive the detection and recognition tasks., Comment: 52 pages, 13 figures

Virtual reality is widely adopted for applications such as training, education, and collaboration. The construction industry, known for its complex projects and numerous personnel involved, relies heavily on effective collaboration. Setting up a real-world construction site for experiments can be expensive and time-consuming, whereas conducting experiments in VR is relatively low-cost, scalable, and efficient. We propose Col-Con, a virtual reality simulation testbed for exploring collaborative behaviors in construction. Col-Con is a multi-user testbed that supports users in completing tasks collaboratively. Additionally, Col-Con provides immersive and realistic simulated construction scenes, where real-time voice communication, along with synchronized transformations, animations, sounds, and interactions, enhances the collaborative experience. As a showcase, we implemented a pipe installation construction task based on Col-Con. A user study demonstrated that Col-Con excels in usability, and participants reported a strong sense of immersion and collaboration. We envision that Col-Con will facilitate research on exploring virtual reality-based collaborative behaviors in construction.

Patient experience and care quality are crucial for a hospital's sustainability and reputation. The analysis of patient feedback offers valuable insight into patient satisfaction and outcomes. However, the unstructured nature of these comments poses challenges for traditional machine learning methods following a supervised learning paradigm. This is due to the unavailability of labeled data and the nuances these texts encompass. This research explores leveraging Large Language Models (LLMs) in conducting Multi-label Text Classification (MLTC) of inpatient comments shared after a stay in the hospital. GPT-4 Turbo was leveraged to conduct the classification. However, given the sensitive nature of patients' comments, a security layer is introduced before feeding the data to the LLM through a Protected Health Information (PHI) detection framework, which ensures patients' de-identification. Additionally, using the prompt engineering framework, zero-shot learning, in-context learning, and chain-of-thought prompting were experimented with. Results demonstrate that GPT-4 Turbo, whether following a zero-shot or few-shot setting, outperforms traditional methods and Pre-trained Language Models (PLMs) and achieves the highest overall performance with an F1-score of 76.12% and a weighted F1-score of 73.61% followed closely by the few-shot learning results. Subsequently, the results' association with other patient experience structured variables (e.g., rating) was conducted. The study enhances MLTC through the application of LLMs, offering healthcare practitioners an efficient method to gain deeper insights into patient feedback and deliver prompt, appropriate responses.

A likelihood analysis of the observables in deeply virtual exclusive photoproduction off a proton target, $ep \rightarrow e' p' \gamma'$, is presented. Two processes contribute to the reaction: deeply virtual Compton scattering, where the photon is produced at the proton vertex, and the Bether-Heitler process, where the photon is radiated from the electron. We consider the unpolarized process for which the largest amount of data with all the kinematic dependences are available from corresponding datasets with unpolarized beams and unpolarized targets from Jefferson Lab. We provide and use a method which derives a joint likelihood of the Compton form factors, which parametrize the deeply virtual Compton scattering amplitude in QCD, for each observed combination of the kinematic variables defining the reaction. The unpolarized twist-two cross section likelihood fully constrains only three of the Compton form factors (CFFs). The impact of the twist-three corrections to the analysis is also explored. The derived likelihoods are explored using Markov chain Monte Carlo (MCMC) methods. Using our proposed method we derive CFF error bars and covariances. Additionally, we explore methods which may reduce the magnitude of error bars/contours in the future., Comment: 22 pages, 11 figures

Optical detection of magnetic resonance using quantum spin sensors (QSS) provides a spatially local and sensitive technique to probe spin dynamics in magnets. However, its utility as a probe of antiferromagnetic resonance (AFMR), wherein the characteristic resonant frequencies substantially exceed the QSS probing frequency, remains an open question. Here, using the nitrogen-vacancy center in diamond as a QSS, we report the first experimental demonstration of optically detected AFMR in layered van der Waals antiferromagnets up to frequencies of 24 GHz, significantly higher than the sensor's spin resonance frequency of 2.87 GHz. To achieve this, we leverage the enhancement of the QSS spin relaxation rate due to low-frequency magnetic field fluctuations that arise from collective late-time dynamics of finite-wavevector magnons excited by the driven uniform AFMR mode. Using optically detected AFMR, we first characterize the temperature and magnetic field dependence on the AFMR modes, which shed light on the intrinsic exchange fields and magnetic anisotropies. Second, we exploit the highly localized sensitivity of the QSS to demonstrate efficient magnon transport over tens of micrometers. Finally, we find that optical detection efficiency in fact increases with increasing frequency, enabling broadband detection of magnetization dynamics. Our work showcases the dual capabilities of QSS as detectors of both high frequency magnetization dynamics and magnon transport, paving the way for understanding and controlling magnetism in Neel states in antiferromagnets.

Generalization to novel object configurations and instances across diverse tasks and environments is a critical challenge in robotics. Keypoint-based representations have been proven effective as a succinct representation for capturing essential object features, and for establishing a reference frame in action prediction, enabling data-efficient learning of robot skills. However, their manual design nature and reliance on additional human labels limit their scalability. In this paper, we propose KALM, a framework that leverages large pre-trained vision-language models (LMs) to automatically generate task-relevant and cross-instance consistent keypoints. KALM distills robust and consistent keypoints across views and objects by generating proposals using LMs and verifies them against a small set of robot demonstration data. Based on the generated keypoints, we can train keypoint-conditioned policy models that predict actions in keypoint-centric frames, enabling robots to generalize effectively across varying object poses, camera views, and object instances with similar functional shapes. Our method demonstrates strong performance in the real world, adapting to different tasks and environments from only a handful of demonstrations while requiring no additional labels. Website: https://kalm-il.github.io/, Comment: CoRL LangRob Workshop, 2024

Broadly intelligent agents should form task-specific abstractions that selectively expose the essential elements of a task, while abstracting away the complexity of the raw sensorimotor space. In this work, we present Neuro-Symbolic Predicates, a first-order abstraction language that combines the strengths of symbolic and neural knowledge representations. We outline an online algorithm for inventing such predicates and learning abstract world models. We compare our approach to hierarchical reinforcement learning, vision-language model planning, and symbolic predicate invention approaches, on both in- and out-of-distribution tasks across five simulated robotic domains. Results show that our approach offers better sample complexity, stronger out-of-distribution generalization, and improved interpretability., Comment: In submission

Optical clocks have improved their frequency stability and estimated accuracy by more than two orders of magnitude over the best caesium microwave clocks that realise the SI second. Accordingly, an optical redefinition of the second has been widely discussed, prompting a need for the consistency of optical clocks to be verified worldwide. While satellite frequency links are sufficient to compare microwave clocks, a suitable method for comparing high-performance optical clocks over intercontinental distances is missing. Furthermore, remote comparisons over frequency links face fractional uncertainties of a few $10^{-18}$ due to imprecise knowledge of each clock's relativistic redshift, which stems from uncertainty in the geopotential determined at each distant location. Here, we report a landmark campaign towards the era of optical clocks, where, for the first time, state-of-the-art transportable optical clocks from Japan and Europe are brought together to demonstrate international comparisons that require neither a high-performance frequency link nor information on the geopotential difference between remote sites. Conversely, the reproducibility of the clocks after being transported between countries was sufficient to determine geopotential height offsets at the level of 4 cm. Our campaign paves the way for redefining the SI second and has a significant impact on various applications, including tests of general relativity, geodetic sensing for geosciences, precise navigation, and future timing networks., Comment: 29 pages, 5 figures

There is mounting evidence that blazars are the sources of part of the very-high-energy astrophysical neutrino flux detected by IceCube. In particular, there have been several spatial and temporal coincidences of individual IceCube neutrino events with flaring blazars, the most prominent of them being IceCube-170922A, coincident with a multi-wavelength flare of TXS~0506+056. Motivated by this, we used the time-dependent lepto-hadronic code OneHaLe to model the spectral energy distributions and light curves of a sample of bright $\gamma$-ray flares of blazars detected by Fermi-LAT, for which Kreter et al. (2020) provided calorimetric estimates of the expected neutrino detection rates. Flares were modelled with temporal changes of the proton injection spectra. Our analysis shows that the calorimetric approach overestimates the increase in neutrino production by a factor of typically $\sim 10$ if the $\gamma$-ray emission is dominated by proton-synchrotron radiation., Comment: 11 pages; Accepted for publication in ApJ

Permutation puzzles, such as the Rubik's cube and the 15 puzzle, are enjoyed by the general public and mathematicians alike. Here we introduce quantum versions of permutation puzzles where the pieces of the puzzles are replaced with indistinguishable quantum particles. The moves in the puzzle are achieved by swapping or permuting the particles. We show that simply permuting the particles can be mapped to a classical permutation puzzle, even though the identical particles are entangled. However, we obtain a genuine quantum puzzle by adding a quantum move-the square root of SWAP. The resulting puzzle cannot be mapped to a classical permutation puzzle. We focus predominately on the quantization of the $2\times 2$ slide puzzle and briefly treat the $2\times 2 \times 1$ Rubik's cube., Comment: 7 Pages + appendices, 4 Figures

Sherpa is a general-purpose Monte Carlo event generator for the simulation of particle collisions in high-energy collider experiments. We summarise new developments, essential features, and ongoing improvements within the Sherpa 3 release series. Physics improvements include higher-order electroweak corrections, simulations of photoproduction and hard diffraction at NLO QCD, heavy-flavour matching in NLO multijet merging, spin-polarised cross section calculations, and a new model of colour reconnections. In addition, the modelling of hadronisation, the underlying event and QED effects in both production and decay has been improved, and the overall event generation efficiency has been enhanced., Comment: We proudly present the new major release of the Sherpa MC generator

Trust and interpretability are crucial for the use of Artificial Intelligence (AI) in scientific research, but current models often operate as black boxes offering limited transparency and justifications for their outputs. We introduce AiSciVision, a framework that specializes Large Multimodal Models (LMMs) into interactive research partners and classification models for image classification tasks in niche scientific domains. Our framework uses two key components: (1) Visual Retrieval-Augmented Generation (VisRAG) and (2) domain-specific tools utilized in an agentic workflow. To classify a target image, AiSciVision first retrieves the most similar positive and negative labeled images as context for the LMM. Then the LMM agent actively selects and applies tools to manipulate and inspect the target image over multiple rounds, refining its analysis before making a final prediction. These VisRAG and tooling components are designed to mirror the processes of domain experts, as humans often compare new data to similar examples and use specialized tools to manipulate and inspect images before arriving at a conclusion. Each inference produces both a prediction and a natural language transcript detailing the reasoning and tool usage that led to the prediction. We evaluate AiSciVision on three real-world scientific image classification datasets: detecting the presence of aquaculture ponds, diseased eelgrass, and solar panels. Across these datasets, our method outperforms fully supervised models in low and full-labeled data settings. AiSciVision is actively deployed in real-world use, specifically for aquaculture research, through a dedicated web application that displays and allows the expert users to converse with the transcripts. This work represents a crucial step toward AI systems that are both interpretable and effective, advancing their use in scientific research and scientific discovery.

While interpretability research has shed light on some internal algorithms utilized by transformer-based LLMs, reasoning in natural language, with its deep contextuality and ambiguity, defies easy categorization. As a result, formulating clear and motivating questions for circuit analysis that rely on well-defined in-domain and out-of-domain examples required for causal interventions is challenging. Although significant work has investigated circuits for specific tasks, such as indirect object identification (IOI), deciphering natural language reasoning through circuits remains difficult due to its inherent complexity. In this work, we take initial steps to characterize causal reasoning in LLMs by analyzing clear-cut cause-and-effect sentences like "I opened an umbrella because it started raining," where causal interventions may be possible through carefully crafted scenarios using GPT-2 small. Our findings indicate that causal syntax is localized within the first 2-3 layers, while certain heads in later layers exhibit heightened sensitivity to nonsensical variations of causal sentences. This suggests that models may infer reasoning by (1) detecting syntactic cues and (2) isolating distinct heads in the final layers that focus on semantic relationships., Comment: 12 pages

Audio production style transfer is the task of processing an input to impart stylistic elements from a reference recording. Existing approaches often train a neural network to estimate control parameters for a set of audio effects. However, these approaches are limited in that they can only control a fixed set of effects, where the effects must be differentiable or otherwise employ specialized training techniques. In this work, we introduce ST-ITO, Style Transfer with Inference-Time Optimization, an approach that instead searches the parameter space of an audio effect chain at inference. This method enables control of arbitrary audio effect chains, including unseen and non-differentiable effects. Our approach employs a learned metric of audio production style, which we train through a simple and scalable self-supervised pretraining strategy, along with a gradient-free optimizer. Due to the limited existing evaluation methods for audio production style transfer, we introduce a multi-part benchmark to evaluate audio production style metrics and style transfer systems. This evaluation demonstrates that our audio representation better captures attributes related to audio production and enables expressive style transfer via control of arbitrary audio effects., Comment: Accepted to ISMIR 2024. Code available https://github.com/csteinmetz1/st-ito

Nonlinear optical materials are essential for the development of both nonlinear and quantum optics and have advanced recently from bulk crystals to integrated material platforms. In this Perspective, we provide an overview of the emerging InGaP $\chi^{(2)}$ nonlinear integrated photonics platform and its experimental achievements. With its exceptional $\chi^{(2)}$ nonlinearity and low optical losses, the epitaxial InGaP platform significantly enhances a wide range of second-order nonlinear optical effects, from second-harmonic generation to entangled photon pair sources, achieving efficiencies several orders of magnitude beyond the current state of the art. Moreover, the InGaP platform enables quantum nonlinear optics at the few- and single-photon levels via passive nonlinearities, which has broad implications for quantum information processing and quantum networking. We also examine the current limitations of the InGaP platform and propose potential solutions to fully unlock its capabilities.

Multiple importance sampling (MIS) is an indispensable tool in rendering that constructs robust sampling strategies by combining the respective strengths of individual distributions. Its efficiency can be greatly improved by carefully selecting the number of samples drawn from each distribution, but automating this process remains a challenging problem. Existing works are mostly limited to mixture sampling, in which only a single sample is drawn in total, and the works that do investigate multi-sample MIS only optimize the sample counts at a per-pixel level, which cannot account for variations beyond the first bounce. Recent work on Russian roulette and splitting has demonstrated how fixed-point schemes can be used to spatially vary sample counts to optimize image efficiency but is limited to choosing the same number of samples across all sampling strategies. Our work proposes a highly flexible sample allocation strategy that bridges the gap between these areas of work. We show how to iteratively optimize the sample counts to maximize the efficiency of the rendered image using a lightweight data structure, which allows us to make local and individual decisions per technique. We demonstrate the benefits of our approach in two applications, path guiding and bidirectional path tracing, in both of which we achieve consistent and substantial speedups over the respective previous state-of-the-art., Comment: 18 pages, 13 figures, to be published in SIGGRAPH Asia 2024 Conference Papers

Establishing a unified theory of cognition has been a major goal of psychology. While there have been previous attempts to instantiate such theories by building computational models, we currently do not have one model that captures the human mind in its entirety. Here we introduce Centaur, a computational model that can predict and simulate human behavior in any experiment expressible in natural language. We derived Centaur by finetuning a state-of-the-art language model on a novel, large-scale data set called Psych-101. Psych-101 reaches an unprecedented scale, covering trial-by-trial data from over 60,000 participants performing over 10,000,000 choices in 160 experiments. Centaur not only captures the behavior of held-out participants better than existing cognitive models, but also generalizes to new cover stories, structural task modifications, and entirely new domains. Furthermore, we find that the model's internal representations become more aligned with human neural activity after finetuning. Taken together, Centaur is the first real candidate for a unified model of human cognition. We anticipate that it will have a disruptive impact on the cognitive sciences, challenging the existing paradigm for developing computational models.

We propose a novel reformulation of the Vlasov-Amp{\`e}re equations for plasmas. This reformulation exposes discrete symmetries to achieve simultaneous conservation of mass, momentum, and energy; preservation of Gauss's law involution; positivity of the distribution function; and quasi-neutral asymptotics. Our approach relies on transforming variables and coordinates, leading to a coupled system of a modified Vlasov equation and the associated moment-field equations. The modified Vlasov equation evolves a conditional distribution function that excludes information on mass, momentum, and energy densities. The mass, momentum, and energy density, in turn, are evolved using moment equations, in which discrete symmetries, conservation laws, and involution constraints are enforced. The reformulation is compatible with a recent slow-manifold reduction technique, which separates the fast electron time scales and simplifies handling the asymptotic quasi-neutrality limit within the easier to solve moment-field subsystem. Using this new formulation, we develop a numerical approach for the reduced 1D1V subsystem that, for the first time, simultaneously satisfies the important physical constraints while preserving the quasi-neutrality asymptotic limit. The advantages of our approach are demonstrated through canonical electrostatic test problems, including the multiscale ion acoustic shock wave problem., Comment: 33 pages, 18 figures

Energy arbitrage is one of the most profitable sources of income for battery operators, generating revenues by buying and selling electricity at different prices. Forecasting these revenues is challenging due to the inherent uncertainty of electricity prices. Deep reinforcement learning (DRL) emerged in recent years as a promising tool, able to cope with uncertainty by training on large quantities of historical data. However, without access to future electricity prices, DRL agents can only react to the currently observed price and not learn to plan battery dispatch. Therefore, in this study, we combine DRL with time-series forecasting methods from deep learning to enhance the performance on energy arbitrage. We conduct a case study using price data from Alberta, Canada that is characterized by irregular price spikes and highly non-stationary. This data is challenging to forecast even when state-of-the-art deep learning models consisting of convolutional layers, recurrent layers, and attention modules are deployed. Our results show that energy arbitrage with DRL-enabled battery control still significantly benefits from these imperfect predictions, but only if predictors for several horizons are combined. Grouping multiple predictions for the next 24-hour window, accumulated rewards increased by 60% for deep Q-networks (DQN) compared to the experiments without forecasts. We hypothesize that multiple predictors, despite their imperfections, convey useful information regarding the future development of electricity prices through a "majority vote" principle, enabling the DRL agent to learn more profitable control policies., Comment: Accepted for publication at the 18th ASME International Conference on Energy Sustainability

The relationship between magnetic activity and Rossby number is one way through which stellar dynamos can be understood. Using measured rotation rates and X-ray to bolometric luminosity ratios of an ensemble of stars, we derive empirical convective turnover times based on recent observations and re-evaluate the X-ray activity-Rossby number relationship. In doing so, we find a sharp rise in the convective turnover time for stars in the mass range of $0.35-0.4\ \rm M_{\odot}$, associated with the onset of a fully convective internal stellar structure. Using $\texttt{MESA}$ stellar evolution models, we infer the location of dynamo action implied by the empirical convective turnover time. The empirical convective turnover time is found to be indicative of dynamo action deep within the convective envelope in stars with masses $0.1-1.2\ \rm M_{\odot}$, crossing the fully convective boundary. Our results corroborate past works suggesting that partially and fully convective stars follow the same activity-Rossby relation, possibly owing to similar dynamo mechanisms. Our stellar models also give insight into the dynamo mechanism. We find that empirically determined convective turnover times correlate with properties of the deep stellar interior. These findings are in agreement with global dynamo models that see a reservoir of magnetic flux accumulate deep in the convection zone before buoyantly rising to the surface., Comment: 9 Figures and 2 Tables

Purpose: Low-field MRI has been assumed to be implant-friendly based on limited studies. However, RF-induced heating due to an implant is a complex resonance phenomenon, highly dependent on the implant's configurations and the applied RF frequencies. This study aims to evaluate the RF heating of DBS implants during MRI at low-field strengths compared to higher field 1.5 T MRI. Methods: A commercial deep brain stimulation (DBS) implant was used in full system as well as lead only configurations to evaluate and compare RF heating during MR imaging at 0.55 T and 1.5 T. The transfer function of the device at both configurations was measured and validated at each of the frequencies, which was then used for the in vivo prediction of RF heating for realistic DBS configurations at head, chest and abdomen imaging landmarks. Results: For the lead only case, the RF heating due to the DBS was substantially smaller during imaging at 0.55 T compared to that at 1.5 T. However, for the full DBS system (longer implant), the RF heating at 0.55 T was comparable to and for some cases even higher than that at 1.5 T, reaching a level that poses risk of tissue damage in patients. Conclusions: While RF heating generally tends to be lower at low-field MRI, the case with longer implanted leads demands extra caution, due to the higher possibility of matching resonant condition at low-field-strength frequencies. Thus, specific risk evaluation for each implant and configuration is required rather than assuming that lower field strength imaging is safer.

GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50\% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models. In line with our commitment to building AI safely and consistent with our voluntary commitments to the White House, we are sharing the GPT-4o System Card, which includes our Preparedness Framework evaluations. In this System Card, we provide a detailed look at GPT-4o's capabilities, limitations, and safety evaluations across multiple categories, focusing on speech-to-speech while also evaluating text and image capabilities, and measures we've implemented to ensure the model is safe and aligned. We also include third-party assessments on dangerous capabilities, as well as discussion of potential societal impacts of GPT-4o's text and vision capabilities.

Metalenses, advanced nanostructured alternatives to conventional lenses, significantly enhance the compactness and functionality of optical systems. Despite progress in monochromatic applications, scaling metalenses to centimeter-sized apertures for broadband or multiwavelength use remains challenging due to limitations in achieving the necessary group delay (GD) with current materials. In this study, we introduce a novel multiwavelength, polarization-insensitive metalens design operating in the near-infrared (NIR). Our approach employs multiple Huygens' metasurface layers, each optimized to modulate a specific wavelength while maintaining high transmittance and minimal phase disturbance at other wavelengths. We demonstrate a metalens operating at 2000 and 2340 nm with a numerical aperture (NA) of 0.11. In simulation, the metalens achieves a normalized modulation transfer function (MTF) that is nearly exactly diffraction-limited. The absolute focusing efficiencies are 65% and 56%, corresponding to relative efficiencies of 76% and 65% compared to an ideal lens of the same dimensions. The meta-atoms are designed using an inverse shape-optimization method that ensures polarization insensitivity and high tolerance to layer misalignment. The proposed design is compatible with large-area nanofabrication techniques, allowing metasurface layers to be fabricated individually and assembled into the final device. This innovative approach is also generalizable to any arbitrary multiwavelength phase profile beyond that of simple lensing., Comment: 9 pages, 5 figures

Active Galactic Nuclei (AGN) are objects located in the heart of galaxies which emit powerful and complex radiation across the electromagnetic spectrum. Understanding AGN has become a topic of interest due to their importance in galactic evolution and their ability to act as a probe to the distant Universe. Within the next few years, wide-field surveys such as the Legacy Survey of Space and Time (LSST) at the Rubin Vera Observatory are expected to increase the number of known AGN to $\mathcal{O} (10^{7})$ and the number of strongly lensed AGN to $\mathcal{O} (10^{4})$. In this paper we introduce \texttt{Amoeba}: an AGN Model of Optical Emission Beyond steady-state Accretion discs. The goal of \texttt{Amoeba} is to provide a modular and flexible modelling environment for AGN, in which all components can interact with each other. Through this work we describe the framework for major AGN components to vary self-consistently and keep flux distributions to connect these components to spatial dependent processes. We model properties beyond traditional single-component models, such as the reverberation of the corona's bending power law power spectrum through the accretion disc and broad line region (BLR). We simulate obscuration by the dusty torus and differential magnification of the disc and BLR due to microlensing. These features are joined together to create some of the most realistic light curve simulations to date. \texttt{Amoeba} takes a step forward in AGN modelling by joining the accretion disc, BLR, torus, intrinsic signal, and microlensing into a coherent model.

Machine learning (ML) is a revolutionary technology with demonstrable applications across multiple disciplines. Within the Earth science community, ML has been most visible for weather forecasting, producing forecasts that rival modern physics-based models. Given the importance of deepening our understanding and improving predictions of the Earth system on all time scales, efforts are now underway to develop forecasting models into Earth-system models (ESMs), capable of representing all components of the coupled Earth system (or their aggregated behavior) and their response to external changes. Modeling the Earth system is a much more difficult problem than weather forecasting, not least because the model must represent the alternate (e.g., future) coupled states of the system for which there are no historical observations. Given that the physical principles that enable predictions about the response of the Earth system are often not explicitly coded in these ML-based models, demonstrating the credibility of ML-based ESMs thus requires us to build evidence of their consistency with the physical system. To this end, this paper puts forward five recommendations to enhance comprehensive, standardized, and independent evaluation of ML-based ESMs to strengthen their credibility and promote their wider use.

Nonlinear superconducting circuits can be used as amplifiers, transducers, and qubits. Only a handful of superconducting circuits have been analyzed or built, so many high-performing configurations likely remain undiscovered. We seek to catalog this design space by enumerating all superconducting circuits -- up to five nodes in size -- built of capacitors, inductors, and Josephson junctions. Using graph isomorphism, we remove redundant configurations to construct a set of unique circuits. We define the concept of a ``Hamiltonian class'' and sort the resulting circuit Hamiltonians based on the types of variables present and the structure of their coupling. Finally, we search for novel superconducting qubits by explicitly considering all three node circuits, showing how the results of our enumeration can be used as a starting point for circuit design tasks., Comment: 16 pages, 11 figures, 11 tables

Quasars are bright active galactic nuclei powered by the accretion of matter around supermassive black holes at the center of galaxies. Their stochastic brightness variability depends on the physical properties of the accretion disk and black hole. The upcoming Rubin Observatory Legacy Survey of Space and Time (LSST) is expected to observe tens of millions of quasars, so there is a need for efficient techniques like machine learning that can handle the large volume of data. Quasar variability is believed to be driven by an X-ray corona, which is reprocessed by the accretion disk and emitted as UV/optical variability. We are the first to introduce an auto-differentiable simulation of the accretion disk and reprocessing. We use the simulation as a direct component of our neural network to jointly model the driving variability and reprocessing to fit simulated LSST 10-year quasar light curves. The driving variability is reconstructed using a latent stochastic differential equation, a physically motivated, generative deep learning method that can model continuous-time stochastic dynamics. By embedding these physical processes into our network, we achieve a model that is more robust and interpretable. We also use transformers to scale our model to tens of millions of parameters. We demonstrate how our model outperforms a Gaussian process regression baseline and can infer accretion disk parameters and time delays between wavebands, even for out-of-distribution driving signals. Our approach provides a powerful and scalable framework that can be adapted to solve other inverse problems in multivariate time series with irregular sampling., Comment: 31 pages, 19 figures

Thermally driven transitions between ferromagnetic and paramagnetic phases are characterized by critical behavior with divergent susceptibilities, long-range correlations, and spin dynamics that can span kHz to GHz scales as the material approaches the critical temperature $\mathrm{T_c}$, but it has proven technically challenging to probe the relevant length and time scales with most conventional measurement techniques. In this study, we employ scanning nitrogen-vacancy center based magnetometry and relaxometry to reveal the critical behavior of a high-$\mathrm{T_c}$ ferromagnetic oxide near its Curie temperature. Cluster analysis of the measured temperature-dependent nanoscale magnetic textures points to a 3D universality class with a correlation length that diverges near $\mathrm{T_c}$. Meanwhile, the temperature-dependent spin dynamics, measured through all optical relaxometry suggest that the phase transition is in the XY universality class. Our results capture both static and dynamic aspects of critical behavior, providing insights into universal properties that govern phase transitions in magnetic materials.

Topological magnons give rise to possibilities for engineering novel spintronics devices with critical applications in quantum information and computation, due to its symmetry-protected robustness and low dissipation. However, to make reliable and systematic predictions about material realization of topological magnons has been a major challenge, due to the lack of neutron scattering data for most materials. In this work, we significantly advance the symmetry-based approach for identifying topological magnons through developing a fully automated algorithm, utilizing the theory of symmetry indicators, that enables a highly efficient and large-scale search for candidate materials hosting field-induced topological magnons. This progress not only streamlines the discovery process but also expands the scope of materials exploration beyond previous manual or traditional methods, offering a powerful tool for uncovering novel topological phases in magnetic systems. Performing a large-scale search over all 1649 magnetic materials in Bilbao Crystallographic Server with a commensurate magnetic order, we discover 387 candidate materials for topological magnons, significantly expanding the pool of topological magnon materials. We further discuss examples and experimental accessibility of the candidate materials, shedding light on future experimental realizations of topological magnons in magnetic materials., Comment: Main text: 30 pages with 2 tables + 12 figures. Supplementary materials: 945 pages with 900+ tables and 460+ figures

Fully optimized automation of behavioral training protocols for lab animals like rodents has long been a coveted goal for researchers. It is an otherwise labor-intensive and time-consuming process that demands close interaction between the animal and the researcher. In this work, we used a data-driven approach to optimize the way rodents are trained in labs. In pursuit of our goal, we looked at data augmentation, a technique that scales well in data-poor environments. Using data augmentation, we built several artificial rodent models, which in turn would be used to build an efficient and automatic trainer. Then we developed a novel similarity metric based on the action probability distribution to measure the behavioral resemblance of our models to that of real rodents., Comment: 5 pages, 3 figures

One of the outstanding goals of the planetary science community is to measure the present-day atmospheric composition of planets and link this back to formation. As giant planets are formed by accreting gas, ices, and rocks, constraining the relative amounts of these components is critical to understand their formation and evolution. For most known planets, including the Solar System giants, this is difficult as they reside in a temperature regime where only volatile elements (e.g., C, O) can be measured, while refractories (e.g., Fe, Ni) are condensed to deep layers of the atmosphere where they cannot be remotely probed. With temperatures allowing for even rock-forming elements to be in the gas phase, ultra-hot Jupiter atmospheres provide a unique opportunity to simultaneously probe the volatile and refractory content of giant planets. Here we directly measure and obtain bounded constraints on the abundances of volatile C and O as well as refractory Fe and Ni on the ultra-hot giant exoplanet WASP-121b. We find that ice-forming elements are comparatively enriched relative to rock-forming elements, potentially indicating that WASP-121b formed in a volatile-rich environment much farther away from the star than where it is currently located. The simultaneous constraint of ice and rock elements in the atmosphere of WASP-121b provides insights into the composition of giant planets otherwise unattainable from Solar System observations., Comment: 24 pages, 9 figures, 2 tables, Accepted for publication in The Astronomical Journal

We present HIP 8522, a young solar twin with the lowest detected lithium, potentially a field blue straggler or the result of episodic early accretion. Its stellar parameters ($T_{\rm eff} = 5729 \pm 7$ K, $\log g = 4.532 \pm 0.016$ dex, $\rm{[Fe/H]} = 0.005 \pm 0.010$ dex, $v_{t} = 1.08 \pm 0.02$ km s$^{-1}$) and chemical composition were determined via spectroscopic equilibrium using high resolution spectra ($R = 60~000-165~000$). The age of HIP 8522 was estimated to be an upper limit of $<$1 Gyr through isochrone fitting and was further confirmed using chemical clocks. Spectral synthesis of the lithium line at $\sim$6707.8 \AA\ yielded an upper lithium abundance limit of $A(\rm{Li}) <$ 0.8 dex. This value is unusually low for solar twins of similar age, which typically have $A(\rm{Li})$ values ranging from 2.0 to 3.3 dex, suggesting that $\sim$2 dex of lithium is missing. We investigate various scenarios, such as planet engulfment, sub-stellar mergers, and extra mixing. However, two distinct hypotheses provide plausible explanations for the significant depletion of lithium: one suggests that HIP 8522 is a field blue straggler formed by the merger of a close binary, while the other proposes that HIP 8522 experienced early episodic accretion. The young solar twin HIP 8522 presents an exceptional opportunity to rigorously test stellar evolution models and gain crucial insights into the internal mixing mechanisms responsible for the significant destruction of lithium.

The increasing presence of AI-generated content on the internet raises a critical question: What happens when generative machine learning models are pretrained on web-scale datasets containing data created by earlier models? Some authors prophesy $\textit{model collapse}$ under a "$\textit{replace}$" scenario: a sequence of models, the first trained with real data and each later one trained only on synthetic data from its preceding model. In this scenario, models successively degrade. Others see collapse as easily avoidable; in an "$\textit{accumulate}$' scenario, a sequence of models is trained, but each training uses all real and synthetic data generated so far. In this work, we deepen and extend the study of these contrasting scenarios. First, collapse versus avoidance of collapse is studied by comparing the replace and accumulate scenarios on each of three prominent generative modeling settings; we find the same contrast emerges in all three settings. Second, we study a compromise scenario; the available data remains the same as in the accumulate scenario -- but unlike $\textit{accumulate}$ and like $\textit{replace}$, each model is trained using a fixed compute budget; we demonstrate that model test loss on real data is larger than in the $\textit{accumulate}$ scenario, but apparently plateaus, unlike the divergence seen with $\textit{replace}$. Third, we study the relative importance of cardinality and proportion of real data for avoiding model collapse. Surprisingly, we find a non-trivial interaction between real and synthetic data, where the value of synthetic data for reducing test loss depends on the absolute quantity of real data. Our insights are particularly important when forecasting whether future frontier generative models will collapse or thrive, and our results open avenues for empirically and mathematically studying the context-dependent value of synthetic data.

The coupling of the angular jitter of the spacecraft and their sub-assemblies with the optical bench and the telescope into the interferometric length readout will be a major noise source in the LISA mission. We refer to this noise as tilt-to-length (TTL) coupling. It will be reduced directly by realignments, and the residual noise will then be subtracted in post-processing. The success of these mitigation strategies depends on an accurate computation of the TTL coupling coefficients. We present here a thorough analysis of the accuracy of the coefficient estimation under different jitter characteristics, angular readout noise levels, and gravitational wave sources. We analyze in which cases the estimates degrade using two estimators, the common least squares estimator and the instrumental variables estimator. Our investigations show that angular readout noise leads to a bias of the least squares estimator, depending on the TTL coupling coefficients, jitter and readout noise level, while the instrumental variable estimator is not biased. We present an equation that predicts the estimation bias of the least squares method due to angular readout noise.

As gravitational-wave observations expand in scope and detection rate, the data analysis infrastructure must be modernized to accommodate rising computational demands and ensure sustainability. We present a scalable gravitational-wave search pipeline which modernizes the GstLAL pipeline by adapting the core filtering engine to the PyTorch framework, enabling flexible execution on both Central Processing Units (CPUs) and Graphics Processing Units (GPUs). Offline search results on the same 8.8 day stretch of public gravitational-wave data indicate that the GstLAL and the PyTorch adaptation demonstrate comparable search performance, even with float16 precision. Lastly, computational benchmarking results show that the GPU float16 configuration of the PyTorch adaptation executed on an A100 GPU can achieve a speedup factor of up to 169 times compared to GstLAL's performance on a single CPU core., Comment: 8 pages, 4 figures

This paper focuses on a domain expert querying system over databases. It presents a solution designed for a French enterprise interested in offering a natural language interface for its clients. The approach, based on entity enrichment, aims at translating natural language queries into database queries. In this paper, the database is treated through a logical paradigm, suggesting the adaptability of our approach to different database models. The good precision of our method is shown through some preliminary experiments.

In today's world, the significance of explainable AI (XAI) is growing in robotics and point cloud applications, as the lack of transparency in decision-making can pose considerable safety risks, particularly in autonomous systems. As these technologies are integrated into real-world environments, ensuring that model decisions are interpretable and trustworthy is vital for operational reliability and safety assurance. This study explores the implementation of SMILE, a novel explainability method originally designed for deep neural networks, on point cloud-based models. SMILE builds on LIME by incorporating Empirical Cumulative Distribution Function (ECDF) statistical distances, offering enhanced robustness and interpretability, particularly when the Anderson-Darling distance is used. The approach demonstrates superior performance in terms of fidelity loss, R2 scores, and robustness across various kernel widths, perturbation numbers, and clustering configurations. Moreover, this study introduces a stability analysis for point cloud data using the Jaccard index, establishing a new benchmark and baseline for model stability in this field. The study further identifies dataset biases in the classification of the 'person' category, emphasizing the necessity for more comprehensive datasets in safety-critical applications like autonomous driving and robotics. The results underscore the potential of advanced explainability models and highlight areas for future research, including the application of alternative surrogate models and explainability techniques in point cloud data., Comment: 17 pages, 9 figures

We introduce ImmerseDiffusion, an end-to-end generative audio model that produces 3D immersive soundscapes conditioned on the spatial, temporal, and environmental conditions of sound objects. ImmerseDiffusion is trained to generate first-order ambisonics (FOA) audio, which is a conventional spatial audio format comprising four channels that can be rendered to multichannel spatial output. The proposed generative system is composed of a spatial audio codec that maps FOA audio to latent components, a latent diffusion model trained based on various user input types, namely, text prompts, spatial, temporal and environmental acoustic parameters, and optionally a spatial audio and text encoder trained in a Contrastive Language and Audio Pretraining (CLAP) style. We propose metrics to evaluate the quality and spatial adherence of the generated spatial audio. Finally, we assess the model performance in terms of generation quality and spatial conformance, comparing the two proposed modes: ``descriptive", which uses spatial text prompts) and ``parametric", which uses non-spatial text prompts and spatial parameters. Our evaluations demonstrate promising results that are consistent with the user conditions and reflect reliable spatial fidelity., Comment: This work pioneers a Latent Diffusion Model for generating text-prompted ambisonic spatial audio

As electrical generation becomes more distributed and volatile, and loads become more uncertain, controllability of distributed energy resources (DERs), regardless of their ownership status, will be necessary for grid reliability. Grid operators lack direct control over end-users' grid interactions, such as energy usage, but incentives can influence behavior -- for example, an end-user that receives a grid-driven incentive may adjust their consumption or expose relevant control variables in response. A key challenge in studying such incentives is the lack of data about human behavior, which usually motivates strong assumptions, such as distributional assumptions on compliance or rational utility-maximization. In this paper, we propose a general incentive mechanism in the form of a constrained optimization problem -- our approach is distinguished from prior work by modeling human behavior (e.g., reactions to an incentive) as an arbitrary unknown function. We propose feedback-based optimization algorithms to solve this problem that each leverage different amounts of information and/or measurements. We show that each converges to an asymptotically stable incentive with (near)-optimality guarantees given mild assumptions on the problem. Finally, we evaluate our proposed techniques in voltage regulation simulations on standard test beds. We test a variety of settings, including those that break assumptions required for theoretical convergence (e.g., convexity, smoothness) to capture realistic settings. In this evaluation, our proposed algorithms are able to find near-optimal incentives even when the reaction to an incentive is modeled by a theoretically difficult (yet realistic) function., Comment: 28 pages, 10 figures

The Cohn-Umans (FOCS '03) group-theoretic framework for matrix multiplication produces fast matrix multiplication algorithms from three subsets of a finite group $G$ satisfying a simple combinatorial condition (the Triple Product Property). The complexity of such an algorithm then depends on the representation theory of $G$. In this paper we extend the group-theoretic framework to the setting of infinite groups. In particular, this allows us to obtain constructions in Lie groups, with favorable parameters, that are provably impossible in finite groups of Lie type (Blasiak, Cohn, Grochow, Pratt, and Umans, ITCS '23). Previously the Lie group setting was investigated purely as an analogue of the finite group case; a key contribution in this paper is a fully developed framework for obtaining bona fide matrix multiplication algorithms directly from Lie group constructions., Comment: 39 pages