Search

Control of microscopic physical systems is a prerequisite for experimental quantum science and its applications. Neutral atomic and molecular systems can be controlled using tunable scattering resonances. However, the resonant control of effective interactions has so far been limited to the ultracold regime, where quantum effects become manifest. Ultracold temperatures are still out of reach for most hybrid trapped ion-atom systems, a prospective platform for quantum technologies and fundamental research. Here we show that magnetically tunable Feshbach resonances can be used to control inelastic collisions between a single trapped Sr${}^+$ ion and Rb atoms high above the ultracold regime. We measure inelastic collision probabilities and use the results to calibrate a comprehensive theoretical model of ion-atom collisions. The observed collision dynamics show signatures of quantum interference, resulting in the pronounced state and mass dependence of the collision rates in the multiple-partial-wave regime. With our model, we discover multiple measurable Feshbach resonances for magnetic fields from 0 to 400 G, which allow significant enhancement of spin-exchange rates at temperatures as high as 1 mK. Future observation of the predicted resonances should allow precise calibration and control of the short-range dynamics in the ${\text{Sr}^++\text{Rb}}$ collisions under unprecedentedly warm conditions., Comment: 10 pages, 3 figures, and Supplementary information (including 1 supplementary figure)

Domain adaptation is often hampered by exceedingly small target datasets and inaccessible source data. These conditions are prevalent in speech verification, where privacy policies and/or languages with scarce speech resources limit the availability of sufficient data. This paper explored techniques of sourcefree domain adaptation unto a limited target speech dataset for speaker verificationin data-scarce languages. Both language and channel mis-match between source and target were investigated. Fine-tuning methods were evaluated and compared across different sizes of labeled target data. A novel iterative cluster-learn algorithm was studied for unlabeled target datasets.

We report on signatures of classical chaos in ultracold collisions between a trapped ion and a free atom. Using numerical simulations, we show that the scattering dynamics can be highly sensitive to initial conditions for various mass ratios and trapping frequencies, indicating the onset of chaos. We quantify this chaotic dynamics by calculating its fractal dimension. We show that for a trapped $^{88}$Sr$^+$ ion and a free $^{87}$Rb atom chaotic dynamic appears under experimentally relevant conditions, and find its characteristic energy scale. The observation of classical chaos in atom-trapped-ion collisions suggests that signatures of quantum chaos might appear, for example, through a Wigner-Dyson distribution of collisional resonances., Comment: 6+3 pages, 5 figures

Many properties of simple finite dimensional gl(m|n)-modules may be better understood by assigning weight diagrams to the highest weights with respect to a given base of simple roots. In this paper we consider bases that are compatible with the standard Borel subalgebra in $gl(m|n)_0 = gl(m) \times gl(n)$; namely, the bases that differ from the distinguished base $\Sigma^{dist}$ of simple roots by a sequence of odd reflections. We examine the weight diagrams that arise from the highest weights of a simple module $L(\lambda)$ with respect to such bases. Further, we provide combinatorial tools to describe all the weight diagrams of highest weights of $L(\lambda)$ provided only with the the weight diagram of $L(\lambda)$ with respect to distinguished highest weight $\lambda$. Finally, we study the maximal cardinality of incomparable sets of positive odd roots with respect to $\Sigma^{dist}$ which are orthogonal to some highest weight of $L(\lambda)$ with respect to a base as above. We provide explicit formulas for this value, connecting it to the combinatorics of the weight diagrams. Based on this study, we respond to the work of M. Gorelik and Th. Heidersdorf, providing a counterexample to their Tail Conjecture.

Secure aggregation is commonly used in federated learning (FL) to alleviate privacy concerns related to the central aggregator seeing all parameter updates in the clear. Unfortunately, most existing secure aggregation schemes ignore two critical orthogonal research directions that aim to (i) significantly reduce client-server communication and (ii) mitigate the impact of malicious clients. However, both of these additional properties are essential to facilitate cross-device FL with thousands or even millions of (mobile) participants. In this paper, we unite both research directions by introducing ScionFL, the first secure aggregation framework for FL that operates efficiently on quantized inputs and simultaneously provides robustness against malicious clients. Our framework leverages (novel) multi-party computation (MPC) techniques and supports multiple linear (1-bit) quantization schemes, including ones that utilize the randomized Hadamard transform and Kashin's representation. Our theoretical results are supported by extensive evaluations. We show that with no overhead for clients and moderate overhead for the server compared to transferring and processing quantized updates in plaintext, we obtain comparable accuracy for standard FL benchmarks. Moreover, we demonstrate the robustness of our framework against state-of-the-art poisoning attacks., Comment: Published in 2024 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML)

Reactions between pairs of atoms are ubiquitous processes in chemistry and physics. Quantum scattering effects on reactions are only observed at extremely ultracold temperatures, close to the $s$-wave regime, with a small number of partial waves involved. At higher temperatures, the different phases associated with the centrifugal barriers of different partial waves average-out quantum interference to yield semi-classical reaction rates. Here we use quantum-logic to experimentally study resonant charge-exchange reactions between single cold pairs of neutral $^{87}$Rb atoms and optically-inaccessible $^{87}$Rb$^{+}$ ions far above the $s$-wave regime. We find that the measured charge-exchange rate is greatly suppressed with respect to the semi-classical prediction. Our results indicate for the first time that quantum interference persists and effects reaction rates at very high temperatures, at least three orders of magnitude higher than the ultracold $s$-wave regime.

Pairs of free particles cannot form bound states in elastic collision due to momentum and energy conservation. In many ultracold experiments, however, the particles collide in the presence of an external trapping potential which can couple the center-of-mass and relative motions and assist the formation of bound-states. Here, we report on observation of weakly bound molecular states formed between one ultracold $^{87}$Rb atom and a single trapped $^{88}$Sr$^+$ ion in the presence of a linear Paul trap. We show that bound states can form efficiently in binary collisions, and enhance the rate of inelastic processes. By observing electronic spin-exchange rate, we study the dependence of these bound states on the collision energy and magnetic field and extract the average molecular binding energy $E_{\textrm{bind}}=0.7(1)$ mK$\cdot k_B$ and the mean lifetime of the molecule $\tau=0.5(1)\,\mu$s, with good agreement with molecular-dynamics simulations. Our simulations predict a highly unusual power-law distribution of molecular lifetimes with a mean that is dominated by extreme, long-lived, events. The dependence of the molecular properties on the trapping parameters opens new avenues to study and control ultracold collisions., Comment: 17 pages, 12 figures

Over the last two years, governments all over the world have used a variety of containment measures to control the spread of COVID-19, such as contact tracing, social distance regulations, and curfews. Epidemiological simulations are commonly used to assess the impact of those policies before they are implemented in actuality. Unfortunately, their predictive accuracy is hampered by the scarcity of relevant empirical data, concretely detailed social contact graphs. As this data is inherently privacy-critical, there is an urgent need for a method to perform powerful epidemiological simulations on real-world contact graphs without disclosing sensitive information. In this work, we present RIPPLE, a privacy-preserving epidemiological modeling framework that enables the execution of a wide range of standard epidemiological models for any infectious disease on a population's most recent real contact graph while keeping all contact information private locally on the participants' devices. In this regard, we also present PIR-SUM, a novel extension to private information retrieval that allows users to securely download the sum of a set of elements from a database rather than individual elements. Our theoretical constructs are supported by a proof-of-concept implementation in which we show that a 2-week simulation over a population of half a million can be finished in 7 minutes with each participant consuming less than 50 KB of data.

Studies of chemical reactions by a single pair of atoms in a well defined quantum state constitute a corner stone in quantum chemistry. Yet, the number of demonstrated techniques which enable observation and control of a single chemical reaction is handful. Here we propose and demonstrate a new technique to study chemical reactions between an ultracold neutral atom and a cold ion using quantum logic. We experimentally study the release of hyperfine energy in a reaction between an ultracold rubidium atom and isotopes of singly ionized strontium for which we do not have experimental control. We detect the reaction outcome and measure the reaction rate of the chemistry ion by reading the motional state of a logic ion via quantum logic, in a single shot. Our work opens new avenues and extends the toolbox of studying chemical reactions, with existing experimental tools, for all atomic and molecular ions in which direct laser cooling and state detection are unavailable.

The cross section of a given process fundamentally quantifies the probability for that given process to occur. In the quantum regime of low energies, the cross section can vary strongly with collision energy due to quantum effects. Here, we report on a method to directly measure the atom-ion collisional cross section in the energy range of 0.2-12 mK$\cdot$ k$_B$, by shuttling ultracold atoms trapped in an optical-lattice across a radio-frequency trapped ion. In this method, the average number of atom-ion collisions per experiment is below one such that the energy resolution is not limited by the broad (power-law) steady-state atom-ion energy distribution. Here, we estimate that the energy resolution is below 200 $\mu$K$\cdot$k$_B$, limited by drifts in the ion's excess micromotion compensation and can be reduced to the 10's $\mu$K$\cdot$k$_B$ regime. This resolution is one order-of-magnitude better than previous experiments measuring cold atom-ion collisional cross section energy dependence. We used our method to measure the energy dependence of the inelastic collision cross sections of a non-adiabatic Electronic-Excitation-Exchange (EEE) and Spin-Orbit Change (SOC) processes. We found that in the measured energy range, the EEE and SOC cross sections statistically agree with the classical Langevin cross section. This method allows for measuring the cross sections of various inelastic processes and opens up possibilities to search for atom-ion quantum signatures such as shape-resonances.

We present a framework that allows to certify the fairness degree of a model based on an interactive and privacy-preserving test. The framework verifies any trained model, regardless of its training process and architecture. Thus, it allows us to evaluate any deep learning model on multiple fairness definitions empirically. We tackle two scenarios, where either the test data is privately available only to the tester or is publicly known in advance, even to the model creator. We investigate the soundness of the proposed approach using theoretical analysis and present statistical guarantees for the interactive test. Finally, we provide a cryptographic technique to automate fairness testing and certified inference with only black-box access to the model at hand while hiding the participants' sensitive data., Comment: Accepted to AIES-2021

In the last decade, hybrid organic-inorganic halide perovskites have emerged as a new type of semiconductor for photovoltaics and other optoelectronic applications. Unlike standard, tetrahedrally bonded semiconductors (e.g. Si and GaAs), the ionic thermal fluctuations in the halide perovskites (i.e. structural dynamics) are strongly coupled to the electronic dynamics. Therefore, it is crucial to obtain accurate and detailed knowledge about the nature of atomic motions within the crystal. This has proved to be challenging due to low thermal stability and the complex, temperature dependent structural phase sequence of the halide perovskites. Here, these challenges are overcome and a detailed analysis of the mode symmetries is provided in the low-temperature orthorhombic phase of methylammonium-lead iodide. Raman measurements using linearly- and circularly- polarized light at 1.16 eV excitation are combined with density functional perturbation theory (DFPT). By performing an iterative analysis of Raman polarization-orientation dependence and DFPT mode analysis, the crystal orientation is determined. Subsequently, accounting for birefringence effects detected using circularly polarized light excitation, the symmetries of all the observed Raman-active modes at 10 K are assigned.

The holy grail of ion-neutral systems is reaching the s-wave scattering regime. However, most of these systems have a fundamental lower collision energy limit which is higher than this s-wave regime. This limit arises from the time-dependant trapping potential of the ion, the Paul trap. In this work, we studied both theoretically and experimentally, the way the Paul trap parameters affect the energy distribution of an ion that is immersed in a bath of ultra-cold atoms. Heating rates and energy distributions of the ion are calculated for various trap parameters by a molecular dynamics (MD) simulation that takes into account the attractive atom-ion potential. The deviation of the energy distribution from a thermal one is discussed. Using the MD simulation, the heating dynamics for different atom-ion combinations is also investigated. In addition, we performed measurements of the heating rates of a ground-state cooled $\ ^{88}$Sr$^+$ ion that is immersed in an ultra-cold cloud of $\ ^{87}$Rb atoms, over a wide range of trap parameters, and compare our results to the MD simulation. Both the simulation and the experiment reveal no significant change in the heating for different parameters of the trap. However, in the experiment a slightly higher global heating is observed, relative to the simulation.

Atom-ion interactions occur through the electric dipole which is induced by the ion on the neutral atom. In a Langevin collision, in which the atom and ion overcome the centrifugal barrier and reach a short internuclear distance, their internal electronic states deform due to their interaction and can eventually alter. Here we explore the outcome products and the energy released from a single Langevin collision between a single cold $^{88}$Sr$^{+}$ ion initialized in the metastable $4d^2D_{5/2,3/2}$ states, and a cold $^{87}$Rb atom in the $5s^2S_{1/2}$ ground state. We found that the long-lived $D_{5/2}$ and $D_{3/2}$ states quench after roughly three Langevin collisions, transforming the excitation energy into kinetic energy. We identify two types of collisional quenching. One is an Electronic Excitation-Exchange process, during which the ion relaxes to the $S$ state and the atom is excited to the $P$ state, followed by energy release of $\sim$ 3000 K$\cdot$k$_B$. The second is Spin-Orbit Change where the ion relaxes from the higher fine-structure $D_{5/2}$ level to the lower $D_{3/2}$ level releasing $\sim$ 400 K$\cdot$k$_B$ into kinetic motion. These processes are theoretically understood to occur through Landau-Zener avoided crossings between the different molecular potential curves. We also found that these relaxation rates are insensitive to the mutual spin orientation of the ion and atoms. This is explained by the strong inertial Coriolis coupling present in ultracold atom-ion collisions due to the high partial wave involved, which strongly mixes different angular momentum states. This inertial coupling explains the loss of the total electronic angular-momentum which is transferred to the external rotation of nuclei. Our results provide deeper understanding of ultracold atom-ion inelastic collisions and offer additional quantum control tools for the cold chemistry field.

IP headers include a 16-bit ID field. Our work examines the generation of this field in Windows (versions 8 and higher), Linux and Android, and shows that the IP ID field enables remote servers to assign a unique ID to each device and thus be able to identify subsequent transmissions sent from that device. This identification works across all browsers and over network changes. In modern Linux and Android versions, this field leaks a kernel address, thus we also break KASLR. Our work includes reverse-engineering of the Windows IP ID generation code, and a cryptanalysis of this code and of the Linux kernel IP ID generation code. It provides practical techniques to partially extract the key used by each of these algorithms, overcoming different implementation issues, and observing that this key can identify individual devices. We deployed a demo (for Windows) showing that key extraction and machine fingerprinting works in the wild, and tested it from networks around the world., Comment: This is an extended paper. The original paper will appear in Usenix Security 2019