Search

Neural Radiance Fields (NeRFs) have emerged as a groundbreaking paradigm for representing 3D objects and scenes by encoding shape and appearance information into the weights of a neural network. Recent works have shown how such weights can be used as input to frameworks processing them to solve deep learning tasks. Yet, these frameworks can only process NeRFs with a specific, predefined architecture. In this paper, we present the first framework that can ingest NeRFs with multiple architectures and perform inference on architectures unseen at training time. We achieve this goal by training a Graph Meta-Network in a representation learning framework. Moreover, we show how a contrastive objective is conducive to obtaining an architecture-agnostic latent space. In experiments on both MLP-based and tri-planar NeRFs, our approach demonstrates robust performance in classification and retrieval tasks that either matches or exceeds that of existing frameworks constrained to single architectures, thus providing the first architecture-agnostic method to perform tasks on NeRFs by processing their weights., Comment: Under review

Over the past decades, near-Earth spacecraft observations have provided insights into the physics of the bow shock, suggesting that solar wind intrinsic turbulence influences the bow shock dynamics. On the other hand, theoretical studies, based on global numerical simulations, have not yet investigated the global 3D interaction between a turbulent solar wind and a planetary magnetosphere. This paper addresses this gap for the first time by investigating the global dynamics of this interaction, providing new perspectives on the underlying physical processes. We examine how the turbulent nature of the solar wind influences the 3D structure and dynamics of magnetized planetary environments or magnetized Earth-like exoplanets, using the newly developed numerical code Menura. We use the hybrid PIC model Menura to conduct 3D simulations of the turbulent solar wind and its interaction with an Earth-like magnetized planet through global numerical simulations of the magnetosphere and its surroundings. We show that solar wind turbulence globally influences the shape and dynamics of the bow shock, the magnetosheath structures, and the ion foreshock dynamics. We show that a turbulent solar wind disrupts the coherence of foreshock fluctuations, induces large fluctuations on the quasi-perpendicular surface of the bow shock, facilitates the formation of bubble-like structures near the bow shock's nose, and modifies the properties of the magnetosheath region. None of these phenomena occur when comparing with the case in which the solar wind is laminar. The turbulent nature of the solar wind impacts the 3D shape and dynamics of the bow shock, magnetosheath, and ion foreshock region. This influence should be considered when studying solar wind-planet interactions in observations and simulations. We discuss the relevance of our findings for current and future missions launched into the heliosphere.

Neural Radiance Fields (NeRFs) have emerged as a standard framework for representing 3D scenes and objects, introducing a novel data type for information exchange and storage. Concurrently, significant progress has been made in multimodal representation learning for text and image data. This paper explores a novel research direction that aims to connect the NeRF modality with other modalities, similar to established methodologies for images and text. To this end, we propose a simple framework that exploits pre-trained models for NeRF representations alongside multimodal models for text and image processing. Our framework learns a bidirectional mapping between NeRF embeddings and those obtained from corresponding images and text. This mapping unlocks several novel and useful applications, including NeRF zero-shot classification and NeRF retrieval from images or text., Comment: Accepted at CVPRW-INRV 2024

In recent years, Neural Fields (NFs) have emerged as an effective tool for encoding diverse continuous signals such as images, videos, audio, and 3D shapes. When applied to 3D data, NFs offer a solution to the fragmentation and limitations associated with prevalent discrete representations. However, given that NFs are essentially neural networks, it remains unclear whether and how they can be seamlessly integrated into deep learning pipelines for solving downstream tasks. This paper addresses this research problem and introduces nf2vec, a framework capable of generating a compact latent representation for an input NF in a single inference pass. We demonstrate that nf2vec effectively embeds 3D objects represented by the input NFs and showcase how the resulting embeddings can be employed in deep learning pipelines to successfully address various tasks, all while processing exclusively NFs. We test this framework on several NFs used to represent 3D surfaces, such as unsigned/signed distance and occupancy fields. Moreover, we demonstrate the effectiveness of our approach with more complex NFs that encompass both geometry and appearance of 3D objects such as neural radiance fields., Comment: Extended version of the paper "Deep Learning on Implicit Neural Representations of Shapes" that was presented at ICLR 2023. Accepted at TPAMI. arXiv admin note: text overlap with arXiv:2302.05438

In clinical trials, patients may discontinue treatments prematurely, breaking the initial randomization and, thus, challenging inference. Stakeholders in drug development are generally interested in going beyond the Intention-To-Treat (ITT) analysis, which provides valid causal estimates of the effect of treatment assignment but does not inform on the effect of the actual treatment receipt. Our study is motivated by an RCT in oncology, where patients assigned the investigational treatment may discontinue it due to adverse events. We propose adopting a principal stratum strategy and decomposing the overall ITT effect into principal causal effects for groups of patients defined by their potential discontinuation behavior. We first show how to implement a principal stratum strategy to assess causal effects on a survival outcome in the presence of continuous time treatment discontinuation, its advantages, and the conclusions one can draw. Our strategy deals with the time-to-event intermediate variable that may not be defined for patients who would not discontinue; moreover, discontinuation time and the primary endpoint are subject to censoring. We employ a flexible model-based Bayesian approach to tackle these complexities, providing easily interpretable results. We apply this Bayesian principal stratification framework to analyze synthetic data of the motivating oncology trial. We simulate data under different assumptions that reflect real scenarios where patients' behavior depends on critical baseline covariates. Supported by a simulation study, we shed light on the role of covariates in this framework: beyond making structural and parametric assumptions more credible, they lead to more precise inference and can be used to characterize patients' discontinuation behavior, which could help inform clinical practice and future protocols.

Driven by the appealing properties of neural fields for storing and communicating 3D data, the problem of directly processing them to address tasks such as classification and part segmentation has emerged and has been investigated in recent works. Early approaches employ neural fields parameterized by shared networks trained on the whole dataset, achieving good task performance but sacrificing reconstruction quality. To improve the latter, later methods focus on individual neural fields parameterized as large Multi-Layer Perceptrons (MLPs), which are, however, challenging to process due to the high dimensionality of the weight space, intrinsic weight space symmetries, and sensitivity to random initialization. Hence, results turn out significantly inferior to those achieved by processing explicit representations, e.g., point clouds or meshes. In the meantime, hybrid representations, in particular based on tri-planes, have emerged as a more effective and efficient alternative to realize neural fields, but their direct processing has not been investigated yet. In this paper, we show that the tri-plane discrete data structure encodes rich information, which can be effectively processed by standard deep-learning machinery. We define an extensive benchmark covering a diverse set of fields such as occupancy, signed/unsigned distance, and, for the first time, radiance fields. While processing a field with the same reconstruction quality, we achieve task performance far superior to frameworks that process large MLPs and, for the first time, almost on par with architectures handling explicit representations., Comment: Accepted at ICLR 2024

The Earth magnetopause, when sufficiently plane and stationary at a local scale, can be considered as a "quasi-tangential" discontinuity, since the normal component of the magnetic field Bn is typically very small but not zero. Contrary to observations, the "Classic Theory of Discontinuities" (CTD) predicts that rotational and compressional jumps should be mutually exclusive in the general case Bn n.e.t. 0, but allows only one exception: the tangential discontinuity provided that Bn is strictly zero. Here we show that Finite Larmor Radius (FLR) effects play an important role in the quasi-tangential case, whenever the ion Larmor radius is not fully negligible with respect to the magnetopause thickness. By including FLR effects, the results suggest that a rotational discontinuity undergoes a change comparable to the change of a Shear Alfven into a Kinetic Alfven wave when considering linear modes. For this new kind of discontinuity, the co-existence of rotational and compressional variations at the magnetopause does no more imply that this boundary is a strict tangential discontinuity, even in 1D-like regions far from X-lines if any. This result may lead to important consequences concerning the oldest and most basic questions of magnetospheric physics: how can the magnetopause be open, where and when? The role of FLR being established theoretically, the paper then shows that it can be proved experimentally. We make use of MMS data and process them with the most recent available 4 spacecraft tools. First, we present the different processing techniques that we use to estimate spatial derivatives, such as grad(B) and div(P), and the magnetopause normal direction. We point out why this normal direction must be determined with extremely high accuracy to make the conclusions unambiguous. The results obtained by these techniques are presented in a detailed case study and on a statistical basis.

Medical image registration seeks to find an optimal spatial transformation that best aligns the underlying anatomical structures. These problems usually require the optimization of a similarity metric. Swarm Intelligence techniques are very effective and efficient optimization methods. This systematic review focuses on 3D medical image registration using Particle Swarm Optimization

We introduce a method to make inference on the composition of a heterogeneous population using survey data, accounting for the possibility that capture heterogeneity is related to key survey variables. To deal with nonignorable nonresponse, we combine different data sources and propose the use of Fisher's noncentral hypergeometric model in a Bayesian framework. To illustrate the potentialities of our methodology, we focus on a case study aimed at estimating the composition of the population of Italian graduates by their occupational status one year after graduating, stratifying by gender and degree program. We account for the possibility that surveys inquiring about the occupational status of new graduates may have response rates that depend on individuals' employment status, implying the nonignorability of the nonresponse. Our findings show that employed people are generally more inclined to answer the questionnaire. Neglecting the nonresponse bias in such contexts might lead to overestimating the employment rate.

We have observed a significant enhancement in the energy deposition by $25$--$100~\mathrm{GeV}$ photons in a $1~\mathrm{cm}$ thick tungsten crystal oriented along its $\langle 111 \rangle$ lattice axes. At $100~\mathrm{GeV}$, this enhancement, with respect to the value observed without axial alignment, is more than twofold. This effect, together with the measured huge increase in secondary particle generation is ascribed to the acceleration of the electromagnetic shower development by the strong axial electric field. The experimental results have been critically compared with a newly developed Monte Carlo adapted for use with crystals of multi-$X_0$ thickness. The results presented in this paper may prove to be of significant interest for the development of high-performance photon absorbers and highly compact electromagnetic calorimeters and beam dumps for use at the energy and intensity frontiers.

Background: Four-corner arthrodesis, which involves fusing four carpal bones while removing the scaphoid bone, is a standard surgery for the treatment of advanced stages of wrist arthritis. Nowadays, it can be performed using a dorsal approach by fixing a plate to the bones and a new radial approach is in development. To date, there is no consensus on the biomechanically optimal and most reliable surgical construct for four-corner arthrodesis. Methods: To evaluate them biomechanically and thus assist the surgeon in choosing the best implant orientation, radial or dorsal, the two different four-corner arthrodesis surgical constructs were virtually simulated on a 3D finite element model representing all major structures of the wrist. Two different realistic load sets were applied to the model, representing common tasks for the elderly. Findings: Results consistency was assessed by comparing with the literature the force magnitude computed on the carpal bones. The Von Mises stress distribution in the radial and dorsal plates were calculated. Stress concentration was located at the plate-screw interface for both surgical constructs, with a maximum stress value of 413 MPa for the dorsal plate compared to 326 MPa for the radial plate, meaning that the stress levels are more unfavourable in the dorsal approach. Interpretation: Although some bending stress was found in one load case, the radial plate was mechanically more robust in the other load case. Despite some limitations, this study provides, for the first time, quantified evidence that the newly developed radial surgical construct is mechanically as efficient as the dorsal surgical construct.

This study is the first to provide a systematic review of the literature focused on the relationship between digitalization and organizational agility (OA). It applies the bibliographic coupling method to 171 peer-reviewed contributions published by 30 June 2021. It uses the digitalization perspective to investigate the enablers, barriers and benefits of processes aimed at providing firms with the agility required to effectively face increasingly turbulent environments. Three different, though interconnected, thematic clusters are discovered and analysed, respectively focusing on big-data analytic capabilities as crucial drivers of OA, the relationship between digitalization and agility at a supply chain level, and the role of information technology capabilities in improving OA. By adopting a dynamic capabilities perspective, this study overcomes the traditional view, which mainly considers digital capabilities enablers of OA, rather than as possible outcomes. Our findings reveal that, in addition to being complex, the relationship between digitalization and OA has a bidirectional character. This study also identifies extant research gaps and develops 13 original research propositions on possible future research pathways and new managerial solutions.

The investigation of the energy frontier in physics requires novel concepts for future colliders. The idea of a muon collider is very appealing since it would allow to study particle collisions at up to tens of TeV energy, while offering a cleaner experimental environment with respect to hadronic colliders. One key element in the muon collider design is the low-emittance muon production. Recently,the Low EMittance Muon Accelerator (LEMMA) collaboration has explored the muon pair production close to its kinematic threshold by annihilating 45 GeV positrons with electrons in a low Z material target. In this configuration, muons are emerging from the target with a naturally low-emittance. In this paper we describe the performance of a system, to study this production mechanism, that consists in several segmented absorbers with alternating active layers composed of fast Cherenkov detectors together with a muon identification technique based on this detector. Passive layers were made of tungsten. We collected data corresponding to muon and electron beams produced at the H2 line in the North Area of the European Organization for Nuclear Research (CERN) in September 2018.

To support state efforts to close postsecondary attainment equity gaps for students of color, Research for Action (RFA) developed a framework that describes three drivers behind efforts to place racial equity at the center of postsecondary policymaking and practice: awareness, inclusion, and integration. The framework is adapted from Equity in the Center's Race Equity Cycle and is informed by Research for Action's examination of the three states in the first cohort of the Lumina Foundation's Talent, Innovation, and Equity (TIE) states -- Colorado, Tennessee, and Oregon. This report highlights four important factors that affect how state higher education agencies worked to advance racial equity: (1) data availability and use; (2) culture at the system or state agency; (3) state leadership; and (4) statewide advocacy environment. This report provides recommendations for how state agencies can drive awareness, inclusion, and integration across each factor. Recommendations are informed by data on the best practices and lessons learned from the implementation of each state's TIE initiative. While each TIE state developed their own strategies for advancing equity and closing racial attainment gaps, the primary pillars of the work were similar -- state-level awareness, coalition-building, and institutional engagement. The first in a series, this report examines state-level strategies for closing racial equity gaps. The next two reports in this series will explore lessons learned in advancing racial equity through statewide coalition-building and investments in institutional efforts.

Research for Action has studied statewide Promise programs since 2017, with in-depth analysis in four states--Delaware, Nevada, Oregon, and Tennessee. Our research in these states includes an extensive review of legislative and policy documents; 146 interviews with policymakers, institutional leaders, and high school staff; site visits to 8 postsecondary institutions; site visits to 12 high schools; 12 high school student focus groups; and 16 college student focus groups. We also conducted quantitative analysis on student-level data from three of the four states involved in the study. This work is part of a larger study led by Research for Action (RFA) that also includes a policy scan of 21 current statewide programs and case studies by Penn AHEAD of local Promise programs in Maryland, Pennsylvania, Washington, and Wisconsin. We summarize additional results of our analysis of state-level Promise programs at www.researchforaction.org/collegepromise. In this case study we outline the program design, successes, and challenges of Delaware's SEED Scholarship as it balances higher education affordability, access, and success. [For the main report, "Statewide College Promise Programs: Balancing Affordability, Access, and Student Success," see ED603766.]

Research for Action has studied statewide Promise programs since 2017, with in-depth analysis in four states--Delaware, Nevada, Oregon, and Tennessee. Our research in these states includes an extensive review of legislative and policy documents; 146 interviews with policymakers, institutional leaders, and high school staff; site visits to 8 postsecondary institutions; site visits to 12 high schools; 12 high school student focus groups; and 16 college student focus groups. We also conducted quantitative analysis on student-level data from three of the four states involved in the study. This work is part of a larger study led by RFA that also includes a policy scan of 21 current statewide programs and case studies by Penn AHEAD of local Promise programs in Maryland, Pennsylvania, Washington, and Wisconsin. We summarize additional results of our analysis of state-level Promise programs at www.researchforaction.org/collegepromise. In this case study we outline the program design, successes, and challenges of Nevada Promise as it balances higher education affordability, access, and success. [For the main report, "Statewide College Promise Programs: Balancing Affordability, Access, and Student Success," see ED603766.]

State-level postsecondary policymaking is daunting. Policymakers navigate through myriad fiscal, political, and ideological constraints and opportunities as they strive to address pressing state concerns or goals. While researchers and advocates may point to "perfect" policy models, in reality perfect policy does not exist. All policymaking involves tradeoffs between what is optimal and what is possible. The job of state policymakers, therefore, is to design and implement policy that can best advance the interests of a state and its residents given the constraints under which they operate. Decisions about each of the five components in our Statewide College Promise Framework have implications for a program's potential to address equity gaps in affordability, access, and success. As we identify the tradeoffs for each component of the Framework, we also highlight how these decisions can advance or limit equity for a wide range of traditionally underserved students.

In 2018, a test run with muons in the North Area at CERN was performed, running parasitically downstream of the COMPASS spectrometer. The aim of the test was to investigate the elastic interactions of muons on atomic electrons, in an experimental configuration similar to the one proposed by the project MUonE, which plans to perform a very precise measurement of the differential cross-section of the elastic interactions. COMPASS was taking data with a 190 GeV pion beam, stopped in a tungsten beam dump: the muons from these pions decays passed through a setup including a graphite target followed by 10 planes of Si tracker and a BGO crystal electromagnetic calorimeter placed at the end of the tracker. The elastic scattering events were analysed, and compared to expectations from MonteCarlo simulation., Comment: 14 pages, 12 figures

Clinical trials often allow patients in the control arm to switch to the treatment arm if their physical conditions are worse than certain tolerance levels. For instance, treatment switching arises in the Concorde clinical trial, which aims to assess causal effects on the time-to-disease progression or death of immediate versus deferred treatment with zidovudine among patients with asymptomatic HIV infection. The Intention-To-Treat analysis does not measure the effect of the actual receipt of the treatment and ignores the information on treatment switching. Other existing methods reconstruct the outcome a patient would have had if they had not switched under strong assumptions. Departing from the literature, we re-define the problem of treatment switching using principal stratification and focus on causal effects for patients belonging to subpopulations defined by the switching behavior under control. We use a Bayesian approach to inference, taking into account that (i) switching happens in continuous time; (ii) switching time is not defined for patients who never switch in a particular experiment; and (iii) survival time and switching time are subject to censoring. We apply this framework to analyze synthetic data based on the Concorde study. Our data analysis reveals that immediate treatment with zidovudine increases survival time for never switcher and that treatment effects are highly heterogeneous across different types of patients defined by the switching behavior.

The rapid growth of text data has motivated the development of machine-learning based automatic text summarization strategies that concisely capture the essential ideas in a larger text. This study aimed to devise an extractive summarization method for A-133 Single Audits, which assess if recipients of federal grants are compliant with program requirements for use of federal funding. Currently, these voluminous audits must be manually analyzed by officials for oversight, risk management, and prioritization purposes. Automated summarization has the potential to streamline these processes. Analysis focused on the "Findings" section of ~20,000 Single Audits spanning 2016-2018. Following text preprocessing and GloVe embedding, sentence-level k-means clustering was performed to partition sentences by topic and to establish the importance of each sentence. For each audit, key summary sentences were extracted by proximity to cluster centroids. Summaries were judged by non-expert human evaluation and compared to human-generated summaries using the ROUGE metric. Though the goal was to fully automate summarization of A-133 audits, human input was required at various stages due to large variability in audit writing style, content, and context. Examples of human inputs include the number of clusters, the choice to keep or discard certain clusters based on their content relevance, and the definition of a top sentence. Overall, this approach made progress towards automated extractive summaries of A-133 audits, with future work to focus on full automation and improving summary consistency. This work highlights the inherent difficulty and subjective nature of automated summarization in a real-world application., Comment: 8 pages, first version

The muon collider represents one of the most promising solutions for a future machine exploring the high energy frontier, but several challenges due to the 2.2 $\mu$sec muon lifetime at rest have to be carefully considered. The LEMMA project is investigating the possibility of producing low emittance muon/antimuon pairs from the e$^+$e$^-$ annihilation process at threshold energy, resulting in small transverse emittance beams without any additional beam cooling. However most of the measurements available are performed at higher $\sqrt{s}$ values. It is therefore necessary to measure muons production in positron annihilation at threshold energy and compare the experimental results with the predictions in this specific energy regime. Apart from being a topic of physical interest by itself, these near to threshold measurements can have a sizeable impact on the estimation of the ultimate luminosity achievable in a muon collider with the LEMMA injection scheme., Comment: 13 pages, 8 figures