Search

Spintronic nano-synapses and nano-neurons perform complex cognitive computations with high accuracy thanks to their rich, reproducible and controllable magnetization dynamics. These dynamical nanodevices could transform artificial intelligence hardware, provided that they implement state-of-the art deep neural networks. However, there is today no scalable way to connect them in multilayers. Here we show that the flagship nano-components of spintronics, magnetic tunnel junctions, can be connected into multilayer neural networks where they implement both synapses and neurons thanks to their magnetization dynamics, and communicate by processing, transmitting and receiving radio frequency (RF) signals. We build a hardware spintronic neural network composed of nine magnetic tunnel junctions connected in two layers, and show that it natively classifies nonlinearly-separable RF inputs with an accuracy of 97.7%. Using physical simulations, we demonstrate that a large network of nanoscale junctions can achieve state-of the-art identification of drones from their RF transmissions, without digitization, and consuming only a few milliwatts, which is a gain of more than four orders of magnitude in power consumption compared to currently used techniques. This study lays the foundation for deep, dynamical, spintronic neural networks.

Extracting information from radiofrequency (RF) signals using artificial neural networks at low energy cost is a critical need for a wide range of applications from radars to health. These RF inputs are composed of multiples frequencies. Here we show that magnetic tunnel junctions can process analogue RF inputs with multiple frequencies in parallel and perform synaptic operations. Using a backpropagation-free method called extreme learning, we classify noisy images encoded by RF signals, using experimental data from magnetic tunnel junctions functioning as both synapses and neurons. We achieve the same accuracy as an equivalent software neural network. These results are a key step for embedded radiofrequency artificial intelligence., Comment: 9 pages, 5 figures

Petri nets are a classical model of concurrency widely used and studied in formal verification with many applications in modeling and analyzing hardware and software, data bases, and reactive systems. The reachability problem is central since many other problems reduce to reachability questions. The reachability problem is known to be decidable but its complexity is extremely high (non primitive recursive). In 2011, a variant of the reachability problem, called the mutual reachability problem, that consists in deciding if two configurations are mutually reachable was proved to be exponential-space complete. Recently, this problem found several unexpected applications in particular in the theory of population protocols. While the mutual reachability problem is known to be definable in the Preburger arithmetic, the best known upper bound of such a formula was recently proved to be non-elementary (tower). In this paper we provide a way to compile the mutual reachability relation of a Petri net with $d$ counters into a quantifier-free Presburger formula given as a doubly exponential disjunction of $O(d)$ linear constraints of exponential size. We also provide some first results about Presburger formulas encoding bottom configurations.

Cosmic dust models are key ingredients in advancing our understanding of astronomical environments as diverse as interstellar clouds in galaxies, circumstellar envelopes around evolved and young stars, and protoplanetary disks. Such models consist of several dust populations, each with different compositions and size distributions. They may also consider different grain shapes, although most models assume spherical grains. All include a component of silicate dust. The absorption and emission properties of these dust components are calculated from the optical constants of each dust material which have various experimental, phenomenological, and theoretical origins depending on the model. We aim to provide the community with new sets of optical constants for amorphous silicate dust analogues at low temperatures. The analogues consist of four Mg-rich silicate samples of stoichiometry ranging from enstatite to olivine, and of eight samples of Mg and Fe rich silicates with a pyroxene stoichiometry and differing magnesium and iron content. We calculated the optical constants from transmission measurements using the Kramers-Kronig relations, assuming that the grains are small compared to the wavelength and prolate in shape with axis ratios of 1.5 and 2 for the Mg and Fe rich samples, respectively. New optical constants for silicate dust analogues were calculated over the wavelength range from 5 to 800-1000 microns, depending on the sample, and at temperatures of 10, 30, 100, 200, and 300 K.We determined the uncertainties on the derived optical constants based on the assumptions used to calculate them. To facilitate the use of these data in cosmic dust models, we provide optical constants extrapolated outside the measured spectral range into the UV-NIR and mm-cm wavelength ranges, as well as formulae that can be used to interpolate them at any temperature in the range 10 - 300 K.

We propose to couple the flux degree of freedom of one mode with the charge degree of freedom of a second mode in a hybrid superconducting-semiconducting architecture. Nonreciprocity can arise in this architecture in the presence of external static magnetic fields alone. We leverage this property to engineer a passive on-chip gyrator, the fundamental two-port nonreciprocal device which can be used to build other nonreciprocal devices such as circulators. We analytically and numerically investigate how the nonlinearity of the interaction, circuit disorder and parasitic couplings affect the scattering response of the gyrator.

Linear mixed-effects models are widely used in analyzing repeated measures data, including clustered and longitudinal data, where inferences of both fixed effects and variance components are of importance. Unlike the fixed effect inference that has been well studied, inference on the variance components is more challenging due to null value being on the boundary and the nuisance parameters of the fixed effects. Existing methods often require strong distributional assumptions on the random effects and random errors. In this paper, we develop empirical likelihood-based methods for the inference of the variance components in the presence of fixed effects. A nonparametric version of the Wilks' theorem for the proposed empirical likelihood ratio statistics for variance components is derived. We also develop an empirical likelihood test for multiple variance components related to a sequence of correlated outcomes. Simulation studies demonstrate that the proposed methods exhibit better type 1 error control than the commonly used likelihood ratio tests when the Gaussian distributional assumptions of the random effects are violated. We apply the methods to investigate the heritability of physical activity as measured by wearable device in the Australian Twin study and observe that such activity is heritable only in the quantile range from 0.375 to 0.514.

Camera sensors are increasingly being combined with machine learning to perform various tasks such as intelligent surveillance. Due to its computational complexity, most of these machine learning algorithms are offloaded to the cloud for processing. However, users are increasingly concerned about privacy issues such as function creep and malicious usage by third-party cloud providers. To alleviate this, we propose an edge-based filtering stage that removes privacy-sensitive attributes before the sensor data are transmitted to the cloud. We use state-of-the-art image manipulation techniques that leverage disentangled representations to achieve privacy filtering. We define opt-in and opt-out filter operations and evaluate their effectiveness for filtering private attributes from face images. Additionally, we examine the effect of naturally occurring correlations and residual information on filtering. We find the results promising and believe this elicits further research on how image manipulation can be used for privacy preservation., Comment: Accepted to the MLCS workshop at ECML PKDD 2022

The recently discovered layered Kagome metals of composition AV3Sb5 (A = K, Rb, Cs) exhibit a complex interplay among superconductivity, charge density wave order, topologically non-trivial electronic band structure and geometrical frustration. Here, we probe the electronic band structure underlying these exotic correlated electronic states in CsV3Sb5 with quantum oscillation measurements in pulsed fields up to 86 T. The high-field data reveal a sequence of magnetic breakdown orbits that allows the construction of a model for the folded Fermi surface of CsV3Sb5. The dominant features are large triangular Fermi surface sheets that cover almost half of the folded Brillouin zone that have not yet been detected in angle resolved photoemission spectroscopy (ARPES). These sheets display pronounced nesting at the charge density wave (CDW) vectors, which may stabilize the CDW state. The Berry phases of the electron orbits have been deduced from Landau level fan diagrams near the quantum limit without the need for extrapolations, thereby unambiguously establishing the non-trivial topological character of several electron bands in this Kagome lattice superconductor., Comment: 28 pages, 10 figures, 1 table, Supplementary Materials added

Over 60,000 songs are released on Spotify every day, and the competition for the listener's attention is immense. In that regard, the importance of captivating and inviting cover art cannot be underestimated, because it is deeply entangled with a song's character and the artist's identity, and remains one of the most important gateways to lead people to discover music. However, designing cover art is a highly creative, lengthy and sometimes expensive process that can be daunting, especially for non-professional artists. For this reason, we propose a novel deep-learning framework to generate cover art guided by audio features. Inspired by VQGAN-CLIP, our approach is highly flexible because individual components can easily be replaced without the need for any retraining. This paper outlines the architectural details of our models and discusses the optimization challenges that emerge from them. More specifically, we will exploit genetic algorithms to overcome bad local minima and adversarial examples. We find that our framework can generate suitable cover art for most genres, and that the visual features adapt themselves to audio feature changes. Given these results, we believe that our framework paves the road for extensions and more advanced applications in audio-guided visual generation tasks., Comment: 8 pages, 6 figures, 4 tables

A conjecture of Mihail and Vazirani states that the edge expansion of the graph of every $0/1$ polytope is at least one. Any lower bound on the edge expansion gives an upper bound for the mixing time of a random walk on the graph of the polytope. Such random walks are important because they can be used to generate an element from a set of combinatorial objects uniformly at random. A weaker form of the conjecture of Mihail and Vazirani says that the edge expansion of the graph of a $0/1$ polytope in $\mathbb{R}^d$ is greater than 1 over some polynomial function of $d$. This weaker version of the conjecture would suffice for all applications. Our main result is that the edge expansion of the graph of a $\textit{random}$ $0/1$ polytope in $\mathbb{R}^d$ is at least $\frac{1}{12d}$ with high probability.

A set of configurations $H$ is a home-space for a set of configurations $X$ of aPetri net if every configuration reachable from (any configuration in) $X$ can reach (some configuration in) $H$. The semilinear home-space problem for Petri nets asks, given a Petri net and semilinear sets of configurations $X$, $H$, if $H$ is a home-space for $X$. In 1989, David de Frutos Escrig and Colette Johnen proved that the problem is decidable when $X$ is a singleton and $H$ is a finite union of linear sets with the same periods. In this paper, we show that the general (semilinear) problem is decidable. This result is obtained by proving a duality between the reachability problem and the non-home-space problem. In particular, we prove that for any Petri net and any semilinear set of configurations $H$ we can effectively compute a semilinear set $C$ of configurations, called a non-reachability core for $H$, such that for every set $X$ the set $H$ is not a home-space for $X$ if, and only if, $C$ is reachable from $X$. We show that the established relation to the reachability problem yields the Ackermann-completeness of the (semilinear) home-space problem. For this we also show that, given a Petri net with an initial marking, the set of minimal reachable markings can be constructed in Ackermannian time.

This article presents a new Domain Specific Embedded Language (DSEL) dedicated to Software-Defined Radio (SDR). From a set of carefully designed components, it enables to build efficient software digital communication systems, able to take advantage of the parallelism of modern processor architectures, in a straightforward and safe manner for the programmer. In particular, proposed DSEL enables the combination of pipelining and sequence duplication techniques to extract both temporal and spatial parallelism from digital communication systems. We leverage the DSEL capabilities on a real use case: a fully digital transceiver for the widely used DVB-S2 standard designed entirely in software. Through evaluation, we show how proposed software DVB-S2 transceiver is able to get the most from modern, high-end multicore CPU targets.