Search

In this work, we investigate the generation of primordial black holes (PBHs) within the framework of single-field inflationary models and their compatibility with the cosmological history of the Universe. Our results suggest that, depending on the masses of the formed PBHs, single-field inflation models require more than fine-tuning a potential to induce ultra-slow roll; it necessitates a comprehensive understanding of the post-inflationary cosmological evolution. As an explicative example, we introduce a new model, based on a double inflection point and consistent with Cosmic Microwave Background observations, capable of generating sub-solar PBHs, whose merger could be potentially detectable by the LVK experiment., Comment: 19 pages and 13 figures

We present a method to prepare non-classical states of the electromagnetic field in a microwave resonator. It is based on a controlled gate that applies a squeezing operation on a SQUID-terminated resonator conditioned on the state of a dispersively coupled qubit. This controlled-squeeze gate, when combined with Gaussian operations on the resonator, is universal. We explore the use of this tool to map an arbitrary qubit state into a superposition of squeezed states. In particular, we target a bosonic code with well-defined superparity which makes photon losses detectable by nondemolition parity measurements. We analyze the possibility of implementing this using state-of-the-art circuit QED tools and conclude that it is within reach of current technologies., Comment: 5 pages, 2 figures and supplemental material

Biomolecular condensates form by phase separation of biological polymers. The cellular functions of the resulting membraneless organelles are closely linked to their physical properties over a wide range of length- and timescales: From the nanosecond dynamics of individual molecules and their interactions, to the microsecond translational diffusion of molecules in the condensates, to their viscoelastic properties at the mesoscopic scale. However, it has remained unclear how to quantitatively link these properties across scales. Here we address this question by combining single-molecule fluorescence, correlation spectroscopy, microrheology, and large-scale molecular dynamics simulations on different condensates that are formed by complex coacervation and span about two orders of magnitude in viscosity and their dynamics at the molecular scale. Remarkably, we find that the absolute timescale of protein chain dynamics in the dense phases can be quantitatively and accurately related to translational diffusion and condensate viscosities by Rouse theory of polymer solutions including entanglement. The simulations indicate that the observed wide range of dynamics arises from different contact lifetimes between amino acid residues, which in the mean-field description of the polymer model cause differences in the friction acting on the chains. These results suggest that remarkably simple physical principles can relate the mesoscale properties of biomolecular condensates to their dynamics at the nanoscale.

In this work, we studied photon generation due to the Dynamical Casimir Effect (DCE) in a one dimensional (1+1) double superconducting cavity. The cavity consists of two perfectly conducting mirrors and a dielectric membrane of infinitesimal depth that effectively couples two cavities. The total length of the double cavity $L$, the difference in length between the two cavities $\Delta L$, and the electric susceptibility $\chi$ and conductivity $v$ of the dielectric membrane are tunable parameters. All four parameters are treated as independent and are allowed to be tuned at the same time, even with different frequencies. We analyzed the cavity's energy spectra under different conditions, finding a transition between two distinct regimes that is accurately described by $k_c=\sqrt{v/\chi}$. In particular, a lowest energy mode is forbidden in one of the regimes while it is allowed in the other. We compared analytical approximations obtained through the Multiple Scale Analysis method with exact numeric solutions, obtaining the typical results when $\chi$ is not being tuned. However, when the susceptibility $\chi$ is tuned, different behaviours (such as oscillations in the number of photons of a cavity prepared in a vacuum state) might arise if the frequencies and amplitudes of all parameters are adequate. These oscillations can be considered as adiabatic shortcuts where all generated photons are eventually destroyed. Finally, we present an equivalent quantum circuit that would allow to experimentally simulate the DCE under the studied conditions., Comment: 22 pages, 20 figures

Motivated by the quantum description of gauge theories, we study the cosmological effects of relaxing the Hamiltonian and momentum constraints in general relativity and Gauss' law in electromagnetism. We show that the unconstrained theories have new source terms that mimic a pressureless dust and a charge density that only follows geodesics. The source terms may be the simplest explanation for dark matter and generically predict a charged component. We comment that discovery of such terms would rule out inflation and be a direct probe of the initial conditions of the universe., Comment: 6 pages

In 2019, more than 5 million providers cared for one or more children either in their own home or in a child's home. Home-based child care (HBCC) providers are a varied group that includes listed providers and unlisted providers who do and do not receive payment. HBCC is especially prevalent in communities of color, communities with high concentrations of people from immigrant backgrounds, areas of concentrated poverty, and rural communities. Yet, research on HBCC lags behind research on center-based child care and early education (CCEE), and the least is known about unlisted providers who do not appear on state or national provider lists and work outside the formal systems supporting CCEE programs. This brief focuses on describing the characteristics of the communities in which unlisted HBCC providers live and, for many, where they typically care for children. It uses the 2019 National Survey of Early Care and Education (NSECE), the American Community Survey (ACS), and the Child Opportunity Index (COI) 2.0.

Fermion soliton stars are a consistent model of exotic compact objects which involve a nonlinear interaction between a real scalar field and fermions through a Yukawa term. This interaction results in an effective fermion mass that depends upon the vacuum structure in the scalar potential. In this work we investigate the tidal deformations of fermion soliton stars and compute the corresponding tidal Love numbers for different model parameters. Furthermore, we discuss the existence of approximate universal relations for the electric and magnetic tidal deformabilities of these stars, and compare them with other solutions of general relativity, such as neutron stars or boson stars. These relations for fermion soliton stars are less universal than for neutron stars, but they are sufficiently different from the ordinary neutron star case that a measurement of the electric and magnetic tidal Love numbers (as potentially achievable by next-generation gravitational wave detectors) can be used to disentangle these families of compact objects. Finally, we discuss the conditions for tidal disruption of fermion soliton stars in a binary system and estimate the detectability of the electromagnetic signal associated with such tidal disruption events., Comment: 15 pages, 4 figures. v2: new figure added, matches version accepted in PRD

We consider a theory in which a real scalar field is Yukawa-coupled to a fermion and has a potential with two non-degenerate vacua. If the coupling is sufficiently strong, a collection of N fermions deforms the true vacuum state, creating energetically-favored false-vacuum pockets in which fermions are trapped. We embed this model within General Relativity and prove that it admits self-gravitating compact objects where the scalar field acquires a non-trivial profile due to non-perturbative effects. We discuss some applications of this general mechanism: i) neutron soliton stars in low-energy effective QCD, which naturally happen to have masses around 2 solar masses and radii around 10 km even without neutron interactions; ii) Higgs false-vacuum pockets in and beyond the Standard Model; iii) dark soliton stars in models with a dark sector. In the latter two examples, we find compelling solutions naturally describing centimeter-size compact objects with masses around 10^-6 solar masses, intriguingly in a range compatible with the OGLE+HSC microlensing anomaly. Besides these interesting examples, the mechanism of non-perturbative vacuum scalarization may play a role in various contexts in and beyond the Standard Model, providing a support mechanism for new compact objects that can form in the early universe, can collapse into primordial black holes through accretion past their maximum mass, and serve as dark matter candidates., Comment: 12 pages, 6 figures. v2: matches version accepted in PRD

A transmission line coupled to an externally driven superconducting quantum interference device (SQUID) can exhibit the Dynamical Casimir Effect (DCE). Employing this setup, we quantize the SQUID degrees of freedom and show that it gives rise to a three-body interaction Hamiltonian with the cavity modes. By considering only two interacting modes from the cavities we show that the device can function as an autonomous cooler where the SQUID can be used as a work source to cool down the cavity modes. Moreover, this setup allows for coupling to all modes existing inside the cavities, and we show that by adding two other extra modes to the interaction with the SQUID the cooling effect can be enhanced., Comment: 13 pages, 6 figures

Unsupervised Domain Adaptation (UDA) aims to harness labeled source data to train models for unlabeled target data. Despite extensive research in domains like computer vision and natural language processing, UDA remains underexplored for time series data, which has widespread real-world applications ranging from medicine and manufacturing to earth observation and human activity recognition. Our paper addresses this gap by introducing a comprehensive benchmark for evaluating UDA techniques for time series classification, with a focus on deep learning methods. We provide seven new benchmark datasets covering various domain shifts and temporal dynamics, facilitating fair and standardized UDA method assessments with state of the art neural network backbones (e.g. Inception) for time series data. This benchmark offers insights into the strengths and limitations of the evaluated approaches while preserving the unsupervised nature of domain adaptation, making it directly applicable to practical problems. Our paper serves as a vital resource for researchers and practitioners, advancing domain adaptation solutions for time series data and fostering innovation in this critical field. The implementation code of this benchmark is available at https://github.com/EricssonResearch/UDA-4-TSC.

In 2019, more than 5 million providers cared for one or more children either in their own home or in a child's home. Home-based child care (HBCC) providers are a varied group that includes both listed providers and unlisted providers who do and do not receive payment. HBCC is especially prevalent in communities of color, communities with high concentrations of people from immigrant backgrounds, areas of concentrated poverty, and rural communities. Yet, research on HBCC lags behind research on center-based child care and early education (CCEE), and the least is known about unlisted providers who do not appear on state or national provider lists and work outside the formal systems supporting CCEE programs. Using the 2019 National Survey of Early Care and Education (NSECE), this brief focuses on how unlisted HBCC providers spend their caregiving time, what kinds of caregiving services they provide, and the characteristics of the children they serve.

In 2019, more than 5 million providers cared for one or more children either in their own home or in a child's home. Home-based child care (HBCC) providers are a varied group that includes both listed providers and unlisted providers who do and do not receive payment. HBCC is especially prevalent in communities of color, communities with high concentrations of people from immigrant backgrounds, areas of concentrated poverty, and rural communities. Yet, research on HBCC lags behind research on center-based child care and early education (CCEE), and the least is known about unlisted providers who do not appear on state or national provider lists and work outside the formal systems supporting CCEE programs. Using the 2019 National Survey of Early Care and Education (NSECE), this brief focuses on unlisted providers' caregiving histories, motivations, and recent experiences with professional supports.

In 2019, more than 5 million providers cared for one or more children either in their own home or in a child's home. Home-based child care (HBCC) providers are a varied group that includes both listed providers and unlisted providers who do and do not receive payment. HBCC is especially prevalent in communities of color, communities with high concentrations of people from immigrant backgrounds, areas of concentrated poverty, and rural communities. Yet, research on HBCC lags behind research on center-based child care and early education (CCEE), and the least is known about unlisted providers who do not appear on state or national provider lists and work outside the formal systems supporting CCEE programs. Using the 2019 National Survey of Early Care and Education (NSECE), this brief focuses on the demographic, educational, economic, and health and wellbeing characteristics of unlisted HBCC providers.

The design process of centrifugal compressors requires applying an optimization process which is computationally expensive due to complex analytical equations underlying the compressor's dynamical equations. Although the regression surrogate models could drastically reduce the computational cost of such a process, the major challenge is the scarcity of data for training the surrogate model. Aiming to strategically exploit the labeled samples, we propose the Active-CompDesign framework in which we combine a thermodynamics-based compressor model (i.e., our internal software for compressor design) and Gaussian Process-based surrogate model within a deployable Active Learning (AL) setting. We first conduct experiments in an offline setting and further, extend it to an online AL framework where a real-time interaction with the thermodynamics-based compressor's model allows the deployment in production. ActiveCompDesign shows a significant performance improvement in surrogate modeling by leveraging on uncertainty-based query function of samples within the AL framework with respect to the random selection of data points. Moreover, our framework in production has reduced the total computational time of compressor's design optimization to around 46% faster than relying on the internal thermodynamics-based simulator, achieving the same performance., Comment: Accepted after peer-review at the 1st workshop on Synergy of Scientific and Machine Learning Modeling, SynS & ML ICML, Honolulu, Hawaii, USA. July, 2023. Copyright 2023 by the author(s)

Fermion soliton stars are a motivated model of exotic compact objects in which a nonlinear self-interacting real scalar field couples to a fermion via a Yukawa term, giving rise to an effective fermion mass that depends on the fluid properties. Here we continue our investigation of this model within General Relativity by considering a scalar potential with generic asymmetric vacua. This case provides fermion soliton stars with a parametrically different scaling of the maximum mass relative to the model parameters, showing that the special case of symmetric vacua, in which we recover our previous results, requires fine tuning. In the more generic case studied here the mass and radius of a fermion soliton star are comparable to those of a neutron star for natural model parameters at the GeV scale. Finally, the asymmetric scalar potential inside the star can provide either a positive or a negative effective cosmological constant in the interior, being thus reminiscent of gravastars or anti-de Sitter bubbles, respectively. In the latter case we find the existence of multiple, disconnected, branches of solutions., Comment: 9 pages, 7 figures. v2: matches version accepted in PRD

Shortcuts to adiabaticity (STA) are relevant in the context of quantum systems, particularly regarding their control when they are subjected to time-dependent external conditions. In this paper, we investigate the completion of a nonadiabatic evolution into a shortcut to adiabaticity for a quantum field confined within a one-dimensional cavity containing two movable mirrors. Expanding upon our prior research, we characterize the field's state using two Moore functions that enables us to apply reverse engineering techniques in constructing the STA. Regardless of the initial evolution, we achieve a smooth extension of the Moore functions that implements the STA. This extension facilitates the computation of the mirrors' trajectories based on the aforementioned functions. Additionally, we draw attention to the existence of a comparable problem within nonrelativistic quantum mechanics., Comment: 17 pages, 8 figures

Partnerships between Early Head Start (EHS) programs and child care providers aim to increase access to high quality, comprehensive services that meet the needs of infants and toddlers from families with low incomes. Quality in early care and education matters. In prior research, high quality early care and education was associated with better outcomes for children, particularly children living in poverty. The EHS Family and Child Experiences Survey (Baby FACES) conceptual framework illustrates how multiple dimensions of quality may be associated with infant and toddler development and learning. In this brief, the authors use this framework and adopt a definition of child care quality to include the structural features of classrooms and characteristics of staff as well as the process quality of the interactions and relationships between teachers and children, and parents and teachers. Drawing on data from Baby FACES 2018, this brief fills a gap in knowledge on EHS-CC partnerships by providing a snapshot of quality in partner classrooms--that is, classrooms in child care partner centers that included at least one child enrolled in EHS. [For "Early Head Start Programs, Staff, and Infants/Toddlers and Families Served: Baby FACES 2018 Data Tables. OPRE Report 2021-92," see ED613544.]

A real scalar field coupled to a fermion via a Yukawa term can evade no-go theorems preventing solitonic solutions. For the first time, we study this model within General Relativity without approximations, finding static and spherically symmetric solutions that describe fermion soliton stars. The Yukawa coupling provides an effective mass for the fermion, which is key to the existence of self-gravitating relativistic solutions. We systematically study this novel family of solutions and present their mass-radius diagram and maximum compactness, which is close to (but smaller than) that of the corresponding Schwarzschild photon sphere. Finally, we discuss the ranges of the parameters of the fundamental theory in which the latter might have interesting astrophysical implications, including compact (sub)solar and supermassive fermion soliton stars for a standard gas of degenerate neutrons and electrons, respectively., Comment: 16 pages, 5 figures. v2: matches version accepted in PRD. v3: final plots, matches PRD version

The development of quantum technologies present important challenges such as the need for fast and precise protocols for implementing quantum operations. Shortcuts to adiabaticity (STA) are a powerful tool for achieving these goals, as they enable us to perform an exactly adiabatic evolution in finite time. In this paper we present a shortcut to adiabaticity for the control of an optomechanical cavity with two moving mirrors. Given reference trajectories for the mirrors, we find analytical expressions that give us effective trajectories which implement a STA for the quantum field inside the cavity. We then solve these equations numerically for different reference protocols, such as expansions, contractions and rigid motions; thus confirming the successful implementation of the STA and finding some general features of these effective trajectories., Comment: 13 pages, 9 figures

We study the dynamical Casimir effect in a double superconducting cavity in a circuit quantum electrodynamics architecture. Parameters in the quantum circuit are chosen in such a way the superconducting cavity can mimic a double cavity, formed by two perfectly conducting outer walls and a dielectric one, with arbitrary permittivity separating both halves. We undertake a spectral analysis of the cavity, showing that the spectrum varies significantly depending on the values of the susceptibility of the dielectric mirror and the relative lengths of both cavities. We study the creation of photons when the walls oscillate harmonically with a small amplitude. Furthermore, we explore the possibility of entangling two uncoupled cavities, starting from a symmetric double cavity and having both of its halves become uncoupled at a later given instant. We consider both cases: (i) when the field is initially in a vacuum state and (ii) the situation in which photon creation via the dynamical Casimir effect has already taken place. We show that the cavities become entangled in both cases but, in the latter, the quantum correlation between individual modes can be greatly increased at the cost of diminishing the entanglement between most pairs of modes., Comment: 18 pages, 12 figures

The use of personal data for training machine learning systems comes with a privacy threat and measuring the level of privacy of a model is one of the major challenges in machine learning today. Identifying training data based on a trained model is a standard way of measuring the privacy risks induced by the model. We develop a novel approach to address the problem of membership inference in pattern recognition models, relying on information provided by adversarial examples. The strategy we propose consists of measuring the magnitude of a perturbation necessary to build an adversarial example. Indeed, we argue that this quantity reflects the likelihood of belonging to the training data. Extensive numerical experiments on multivariate data and an array of state-of-the-art target models show that our method performs comparable or even outperforms state-of-the-art strategies, but without requiring any additional training samples.

Shortcuts to adiabaticity constitute a powerful alternative that speed up time-evolution while mimicking adiabatic dynamics. They are also relevant to clarify fundamental questions such as a precise quantification of the third principle of thermodynamics and quantum speed limits. In this letter we describe, for the first time, how to implement shortcuts to adiabaticity in quantum field theory, for the particular case of a massless scalar field inside a cavity with a moving wall, in 1 + 1 dimensions. The approach is based on the known solution to the problem that exploits the conformal symmetry, and the shortcuts take place whenever there is no dynamical Casimir effect. We obtain a fundamental limit for the efficiency of an Otto cycle with the quantum field as a working system, that depends on the maximum velocity that the mirror can attain. We describe possible experimental realizations of the shortcuts using superconducting circuits., Comment: Published version

When combined with high quality early care and education, comprehensive services are intended to support families in their role as caregivers and foster the health and well-being of children. Connecting children and families to these services is a foundational feature of the Head Start and Early Head Start (EHS) models. Partnerships between EHS and other early care and education settings hold promise for expanding access to comprehensive services for infants, toddlers, and their families. Specifically, EHS-Child Care (EHS-CC) Partnerships may extend the reach of comprehensive services to all children enrolled in partnering child care centers and family child care homes (that is, children in partnership and nonpartnership slots). This brief draws on data from the national descriptive study of EHS-CC Partnerships to describe the range of services offered to children in both partnership and nonpartnership slots and their families. The national descriptive study was designed to develop a rich knowledge base about the EHS programs, community-based child care centers, and family child care providers participating in a 2015 federal grants program supporting the development of EHS-CC Partnerships and aiming to increase access to high quality infant-toddler care for low-income families. [For "Working Together for Children and Families: Findings from the National Descriptive Study of Early Head Start-Child Care Partnerships. OPRE Report 2019-16," see ED606742.]

Prior research suggests that partnerships in early care and education have the potential to enhance the quality of care and offer comprehensive services to more children and families. This brief draws on data from the national descriptive study of Early Head Start-Child Care (EHS-CC) Partnerships to describe the activities partnerships engage in to improve the quality of services for infants, toddlers, and their families. The national descriptive study was designed to develop a rich knowledge base about the EHS programs, community-based child care centers, and family child care providers participating in a 2015 federal grants program supporting the development of EHS-CC Partnerships and aiming to increase access to high quality infant-toddler care for low-income families. The national descriptive study is the first to examine partnering EHS programs and child care providers within a national sample. The data presented in this brief provide national estimates of the quality improvement activities they engaged in through their partnerships. Additionally, this brief highlights findings from in-depth case studies of 10 EHS-CC Partnerships to illustrate some of the quality improvement opportunities and challenges they encountered. [For "Working Together for Children and Families: Findings from the National Descriptive Study of Early Head Start-Child Care Partnerships. OPRE Report 2019-16," see ED606742.]

We analyze the efficiency of the quantum Otto cycle applied to a superconducting cavity. We consider its description in terms of a full quantum scalar field in a one-dimensional cavity with a time dependent boundary condition that can be externally controlled to perform and extract work unitarily from the system. We study the performance of this machine when acting as a heat engine as well as a refrigerator. It is shown that, in a non-adiabatic regime, the efficiency of the quantum cycle is affected by the dynamical Casimir effect, that induces a sort of quantum friction that diminishes the efficiency. We also find regions of parameters where the effect is so strong that the machine can no longer function as an engine since the work that would be produced is completely consumed by the quantum friction. However, this effect can be avoided for some particular temporal evolutions of the boundary conditions that do not change the occupation number of the modes in the cavity, leading to a highly improved efficiency., Comment: 11 pages, 11 figures

The Born rule postulates that the probability of measurement in quantum mechanics is related to the squared modulus of the wave function $\psi$. We rearrange the equation for energy eigenfunctions to define the energy as the real part of $\hat{E}\psi/\psi$. For an eigenstate, this definition gives a constant energy eigenvalue. For a general wave function, the energy fluctuates in space and time. We consider a particle in a one-dimensional square well potential in a superposition of two states and average the energy over space and time. We show that, for most cases, such an energy expectation value differs by only a few percent from that calculated using the Born rule. This difference is consistent with experimental tests of the expectation value and suggests that the Born rule may be an approximation of spacetime averaging., Comment: 4 pages, 5 figures

In the presence of spacetime torsion, the momentum components do not commute; therefore, in quantum field theory, summation over the momentum eigenvalues will replace integration over the momentum. In the Einstein--Cartan theory of gravity, in which torsion is coupled to spin, the separation between the eigenvalues increases with the magnitude of the momentum. Consequently, this replacement regularizes divergent integrals in Feynman diagrams with loops by turning them into convergent sums. In this article, we apply torsional regularization to the self-energy of a charged lepton in quantum electrodynamics. We show that this procedure eliminates the ultraviolet divergence. We also show that torsion gives a photon a small nonzero mass, which regularizes the infrared divergence. In the end, we calculate the finite bare masses of the electron, muon, and tau lepton: $0.4329\,\mbox{MeV}$, $90.95\,\mbox{MeV}$, and $1543\,\mbox{MeV}$, respectively. These values constitute about $85\%$ of the observed, re-normalized masses., Comment: 7 pages

Recently, it has been shown that Machine Learning models can leak sensitive information about their training data. This information leakage is exposed through membership and attribute inference attacks. Although many attack strategies have been proposed, little effort has been made to formalize these problems. We present a novel formalism, generalizing membership and attribute inference attack setups previously studied in the literature and connecting them to memorization and generalization. First, we derive a universal bound on the success rate of inference attacks and connect it to the generalization gap of the target model. Second, we study the question of how much sensitive information is stored by the algorithm about its training set and we derive bounds on the mutual information between the sensitive attributes and model parameters. Experimentally, we illustrate the potential of our approach by applying it to both synthetic data and classification tasks on natural images. Finally, we apply our formalism to different attribute inference strategies, with which an adversary is able to recover the identity of writers in the PenDigits dataset., Comment: Published in [Elsevier Neurocomputing](https://doi.org/10.1016/j.neucom.2023.02.058)

Young children's exposure to early math concepts is important in their development, in their confidence in math skills, and in their ability to use math later in life. In 2013, the Heising-Simons Foundation embarked on an innovative path to support children's early math development. The Foundation invited several family engagement providers to develop, test, and integrate research-based early math projects within their existing family engagement programs. The Foundation has since supported two providers--Reach Out and Read (ROR) and the YMCA of Silicon Valley (YMCA)--as they refined their early math projects and scaled them up in new sites. Mathematica conducted an implementation study that examined the scale-up effort during 2018 and 2019. We share the study findings in this brief so program operators, practitioners, and policymakers interested in scaling up early math programs can benefit from the experiences and lessons learned from these two pioneering organizations.

High quality early learning experiences can promote young children's development and help to reduce achievement gaps between children from low-income families and children from more affluent families. Early care and education programs also promote parents' ability to support their children's learning, and allow parents to work or go to school. However, affordable, high quality, child care for infants and toddlers from low-income families is scarce. One strategy for improving access to high quality care for infants and toddlers is to form partnerships at the point of service delivery to build seamless systems of care and promote quality across settings. In 2015, the Administration for Children and Families (ACF) awarded 275 Early Head Start (EHS) Expansion and EHS-Child Care (EHS-CC) Partnership grants. Of these, 250 grantees received funding for EHS-CC Partnerships or funding for both EHS-CC Partnerships and EHS Expansion. The EHS-CC Partnership grants support partnerships between EHS grantees and regulated child care centers and family child care homes serving infants and toddlers from low-income families. The partnerships aim to bring together the best of both programs by combining the high quality, comprehensive, relationship-based child development and family services of EHS with the flexibility of child care and its responsiveness to the social, cultural, and work-support needs of families. To better understand the characteristics of early care and education partnerships, including the EHS-CC Partnerships, the Office of Planning, Research, and Evaluation (OPRE) in ACF commissioned a national descriptive study of EHS-CC Partnerships. Through a contract with Mathematica Policy Research, the national descriptive study provides a rich knowledge base about the characteristics of EHS-CC Partnerships and strategies for implementing partnerships in both center-based child care and family child care homes.

Power consumption data is very useful as it allows to optimize power grids, detect anomalies and prevent failures, on top of being useful for diverse research purposes. However, the use of power consumption data raises significant privacy concerns, as this data usually belongs to clients of a power company. As a solution, we propose a method to generate synthetic power consumption samples that faithfully imitate the originals, but are detached from the clients and their identities. Our method is based on Generative Adversarial Networks (GANs). Our contribution is twofold. First, we focus on the quality of the generated data, which is not a trivial task as no standard evaluation methods are available. Then, we study the privacy guarantees provided to members of the training set of our neural network. As a minimum requirement for privacy, we demand our neural network to be robust to membership inference attacks, as these provide a gateway for further attacks in addition to presenting a privacy threat on their own. We find that there is a compromise to be made between the privacy and the performance provided by the algorithm.