Search

Soft gels, formed via the self-assembly of particulate organic materials, exhibit intricate multi-scale structures that provides them with flexibility and resilience when subjected to external stresses. This work combines molecular simulations and topological data analysis (TDA) to characterize the complex multi-scale structure of soft gels. Our TDA analysis focuses on the use of the Euler characteristic, which is an interpretable and computationally-scalable topological descriptor that is combined with filtration operations to obtain information on the geometric (local) and topological (global) structure of soft gels. We reduce the topological information obtained with TDA using principal component analysis (PCA) and show that this provides an informative low-dimensional representation of gel structure. We use the proposed computational framework to investigate the influence of gel preparation (e.g., quench rate, volume fraction) on soft gel structure and to explore dynamic deformations that emerge under oscillatory shear in various response regimes (linear, nonlinear, and flow). Our analysis identifies specific scales and extents at which hierarchical structures in soft gels are affected; moreover, correlations between structural deformations and mechanical phenomena (such as shear stiffening) are explored. In summary, we show that TDA facilitates the mathematical representation, quantification, and analysis of soft gel structures, extending traditional network analysis methods to capture both local and global organization.

Transformer networks are rapidly becoming SotA in many fields, such as NLP and CV. Similarly to CNN, there is a strong push for deploying Transformer models at the extreme edge, ultimately fitting the tiny power budget and memory footprint of MCUs. However, the early approaches in this direction are mostly ad-hoc, platform, and model-specific. This work aims to enable and optimize the flexible, multi-platform deployment of encoder Tiny Transformers on commercial MCUs. We propose a complete framework to perform end-to-end deployment of Transformer models onto single and multi-core MCUs. Our framework provides an optimized library of kernels to maximize data reuse and avoid unnecessary data marshaling operations into the crucial attention block. A novel MHSA inference schedule, named Fused-Weight Self-Attention, is introduced, fusing the linear projection weights offline to further reduce the number of operations and parameters. Furthermore, to mitigate the memory peak reached by the computation of the attention map, we present a Depth-First Tiling scheme for MHSA. We evaluate our framework on three different MCU classes exploiting ARM and RISC-V ISA, namely the STM32H7, the STM32L4, and GAP9 (RV32IMC-XpulpV2). We reach an average of 4.79x and 2.0x lower latency compared to SotA libraries CMSIS-NN (ARM) and PULP-NN (RISC-V), respectively. Moreover, we show that our MHSA depth-first tiling scheme reduces the memory peak by up to 6.19x, while the fused-weight attention can reduce the runtime by 1.53x, and number of parameters by 25%. We report significant improvements across several Tiny Transformers: for instance, when executing a transformer block for the task of radar-based hand-gesture recognition on GAP9, we achieve a latency of 0.14ms and energy consumption of 4.92 micro-joules, 2.32x lower than the SotA PULP-NN library on the same platform., Comment: Pre-print manuscript submitted for review to the IEEE Transactions on Computers

The emergence of differentiable simulators enabling analytic gradient computation has motivated a new wave of learning algorithms that hold the potential to significantly increase sample efficiency over traditional Reinforcement Learning (RL) methods. While recent research has demonstrated performance gains in scenarios with comparatively smooth dynamics and, thus, smooth optimization landscapes, research on leveraging differentiable simulators for contact-rich scenarios, such as legged locomotion, is scarce. This may be attributed to the discontinuous nature of contact, which introduces several challenges to optimizing with analytic gradients. The purpose of this paper is to determine if analytic gradients can be beneficial even in the face of contact. Our investigation focuses on the effects of different soft and hard contact models on the learning process, examining optimization challenges through the lens of contact simulation. We demonstrate the viability of employing analytic gradients to learn physically plausible locomotion skills with a quadrupedal robot using Short-Horizon Actor-Critic (SHAC), a learning algorithm leveraging analytic gradients, and draw a comparison to a state-of-the-art RL algorithm, Proximal Policy Optimization (PPO), to understand the benefits of analytic gradients.

The coherent production of vector mesons in photon - hadron interactions is considered one of the most promising observables to probe the QCD dynamics at high energies and the transverse spatial distribution of the gluons in the hadron wave function. In this paper, we perform an exploratory study about the dependence of the transverse momentum distributions, $d\sigma/dt$, on the model assumed for the proton density profile. We consider a set of non-Gaussian profiles and include them in the forward dipole - proton scattering amplitude associated with the IP-sat model. We demonstrate that the predictions for $d\sigma/dt$ are similar in the HERA kinematical range, but are very distinct for values of $t$ that will be probed in future colliders. In particular, the presence of diffractive dips in the $t$-distributions associated with the production of $J/\Psi$ and $\rho$ mesons is strongly dependent on the model assumed for the density profile. Similar conclusions are also derived when the non-linear effects on the dipole - target interaction are disregarded. Our results indicate that a future study of the coherent vector meson production at large - $t$ will be useful to constrain the spatial distribution of gluons within the proton., Comment: 11 pages, 5 figures, 1 table. Version to be published in European Physical Journal C

Feasibility of two-line design of Josephson traveling wave parametric amplifier aimed at increase of the allowed pump wave energy and hence the gain growth is analyzed and discussed. Serious restrictions follow from both the cyclic energy transfer of the pump, signal and idler waves in the coupled waveguide lines and the phase mismatch of the waves. Besides, impact of the artificial line discreteness on the phase mismatch is considered as well., Comment: 6 pages, 6 figures. v2: Small mistypes in formulas (6), (12), (33), (36) and (37) have been corrected. No changes have been made based on the comments from any peer-review process

Most 3D Gaussian Splatting (3D-GS) based methods for urban scenes initialize 3D Gaussians directly with 3D LiDAR points, which not only underutilizes LiDAR data capabilities but also overlooks the potential advantages of fusing LiDAR with camera data. In this paper, we design a novel tightly coupled LiDAR-Camera Gaussian Splatting (TCLC-GS) to fully leverage the combined strengths of both LiDAR and camera sensors, enabling rapid, high-quality 3D reconstruction and novel view RGB/depth synthesis. TCLC-GS designs a hybrid explicit (colorized 3D mesh) and implicit (hierarchical octree feature) 3D representation derived from LiDAR-camera data, to enrich the properties of 3D Gaussians for splatting. 3D Gaussian's properties are not only initialized in alignment with the 3D mesh which provides more completed 3D shape and color information, but are also endowed with broader contextual information through retrieved octree implicit features. During the Gaussian Splatting optimization process, the 3D mesh offers dense depth information as supervision, which enhances the training process by learning of a robust geometry. Comprehensive evaluations conducted on the Waymo Open Dataset and nuScenes Dataset validate our method's state-of-the-art (SOTA) performance. Utilizing a single NVIDIA RTX 3090 Ti, our method demonstrates fast training and achieves real-time RGB and depth rendering at 90 FPS in resolution of 1920x1280 (Waymo), and 120 FPS in resolution of 1600x900 (nuScenes) in urban scenarios.

We present the first detection in space of thionylimide (HNSO) toward the Galactic Center molecular cloud G+0.693-0.027, thanks to the superb sensitivity of an ultradeep molecular line survey carried out with the Yebes 40$\,$m and IRAM 30$\,$m telescopes. This molecule is the first species detected in the interstellar medium containing, simultaneously, N, S and O. We have identified numerous $K$$_a$ = 0, 1 and 2 transitions belonging to HNSO covering from $J$$_{\rm up}$ = 2 to $J$$_{\rm up}$ = 10, including several completely unblended features. We derive a molecular column density of $N$ = (8 $\pm$ 1)$\times$10$^{13}$ cm$^{-2}$, yielding a fractional abundance relative to H$_2$ of $\sim$6$\times$10$^{-10}$, which is about $\sim$37 and $\sim$4.8 times less abundant than SO and SO2, respectively. Although there are still many unknowns in the interstellar chemistry of NSO-bearing molecules, we propose that HNSO is likely formed through the reaction of the NSO radical and atomic H on the surface of icy grains, with alternative routes also deserving exploration. Finally, HNSO appears as a promising link between N- , S- and O- interstellar chemistry and its discovery paves the route to the detection of a new family of molecules in space., Comment: Forthcoming paper in The Astrophysical Journal Letters (in press)

Cascades are a common type of machine learning systems in which a large, remote model can be queried if a local model is not able to accurately label a user's data by itself. Serving stacks for large language models (LLMs) increasingly use cascades due to their ability to preserve task performance while dramatically reducing inference costs. However, applying cascade systems in situations where the local model has access to sensitive data constitutes a significant privacy risk for users since such data could be forwarded to the remote model. In this work, we show the feasibility of applying cascade systems in such setups by equipping the local model with privacy-preserving techniques that reduce the risk of leaking private information when querying the remote model. To quantify information leakage in such setups, we introduce two privacy measures. We then propose a system that leverages the recently introduced social learning paradigm in which LLMs collaboratively learn from each other by exchanging natural language. Using this paradigm, we demonstrate on several datasets that our methods minimize the privacy loss while at the same time improving task performance compared to a non-cascade baseline.

Telecom C-band single photons exhibit the lowest attenuation in optical fibers, enabling long-haul quantum-secured communication. However, efficient coupling with optical fibers is crucial for these single photons to be effective carriers in long-distance transmission. In this work, we demonstrate an efficient fiber-coupled single photon source at the telecom C-band using InAs/InP quantum dots coupled to a tapered nanobeam. The tapered nanobeam structure facilitates directional emission that is mode-matched to a lensed fiber, resulting in a collection efficiency of up to 65% from the nanobeam to a single-mode fiber. Using this approach, we demonstrate single photon count rates of 575 $\pm$ 5 Kcps and a single photon purity of $g^2$ (0) = 0.015 $\pm$ 0.003. Additionally, we demonstrate Hong-Ou Mandel interference from the emitted photons with a visibility of 0.84 $\pm$ 0.06. From these measurements, we determine a photon coherence time of 450 $\pm$ 20 ps, a factor of just 8.3 away from the lifetime limit. This work represents an important step towards the development of telecom C-band single-photon sources emitting bright, pure, and indistinguishable photons, which are necessary to realize fiber-based long-distance quantum networks

This paper provides an approach to design control barrier functions (CBFs) using the notion of dynamic safety margins (DSMs). In particular, it is shown that DSMs are CBFs for an augmented system. The proposed approach can handle multiple state and input constraints using the control-sharing property of CBFs. Moreover, it makes no assumption on the relative degree of the constraints. Numerical simulations show that the method outperforms existing DSM-based approaches, while also guaranteeing safety and recursive feasibility., Comment: 7 pages, 2 figures

Several recent papers have investigated the potential of language models as knowledge bases as well as the existence of severe biases when extracting factual knowledge. In this work, we focus on the factual probing performance over unseen prompts from tuning, and using a probabilistic view we show the inherent misalignment between pre-training and downstream tuning objectives in language models for probing knowledge. We hypothesize that simultaneously debiasing these objectives can be the key to generalisation over unseen prompts. We propose an adapter-based framework, UniArk, for generalised and consistent factual knowledge extraction through simple methods without introducing extra parameters. Extensive experiments show that UniArk can significantly improve the model's out-of-domain generalisation as well as consistency under various prompts. Additionally, we construct ParaTrex, a large-scale and diverse dataset for measuring the inconsistency and out-of-domain generation of models. Further, ParaTrex offers a reference method for constructing paraphrased datasets using large language models., Comment: NAACL 2024

The conditions required by quantum matter to modify space-time geometry are explored within the framework of the general theory of relativity. The required characteristics for space-time modification in solid state structures, are met in either (a) massive photon Bose-Einstein condensate in a waveguide, or (b) the massive photons in superconductor's bulk, or (c) the Bose-Einstein condensate of acoustic phonons, or (d) a metal-insulator-topological insulator heterostructure.

State-of-the-art language model fine-tuning techniques, such as Direct Preference Optimization (DPO), restrict user control by hard-coding predefined behaviors into the model. To address this, we propose a novel method, Configurable Safety Tuning (CST), that augments DPO using synthetic preference data to facilitate flexible safety configuration of LLMs at inference time. CST overcomes the constraints of vanilla DPO by introducing a system prompt specifying safety configurations, enabling LLM deployers to disable/enable safety preferences based on their need, just changing the system prompt. Our experimental evaluations indicate that CST successfully manages different safety configurations and retains the original functionality of LLMs, showing it is a robust method for configurable deployment. Data and models available at https://github.com/vicgalle/configurable-safety-tuning

Recent trends are emerging in the use of Large Language Models (LLMs) as autonomous agents that take actions based on the content of the user text prompts. We intend to apply these concepts to the field of Guidance, Navigation, and Control in space, enabling LLMs to have a significant role in the decision-making process for autonomous satellite operations. As a first step towards this goal, we have developed a pure LLM-based solution for the Kerbal Space Program Differential Games (KSPDG) challenge, a public software design competition where participants create autonomous agents for maneuvering satellites involved in non-cooperative space operations, running on the KSP game engine. Our approach leverages prompt engineering, few-shot prompting, and fine-tuning techniques to create an effective LLM-based agent that ranked 2nd in the competition. To the best of our knowledge, this work pioneers the integration of LLM agents into space research. Code is available at https://github.com/ARCLab-MIT/kspdg., Comment: Source code available on Github at: https://github.com/ARCLab-MIT/kspdg

Pretrained language models underpin several AI applications, but their high computational cost for training limits accessibility. Initiatives such as BLOOM and StarCoder aim to democratize access to pretrained models for collaborative community development. However, such existing models face challenges: limited multilingual capabilities, continual pretraining causing catastrophic forgetting, whereas pretraining from scratch is computationally expensive, and compliance with AI safety and development laws. This paper presents Aurora-M, a 15B parameter multilingual open-source model trained on English, Finnish, Hindi, Japanese, Vietnamese, and code. Continually pretrained from StarCoderPlus on 435 billion additional tokens, Aurora-M surpasses 2 trillion tokens in total training token count. It is the first open-source multilingual model fine-tuned on human-reviewed safety instructions, thus aligning its development not only with conventional red-teaming considerations, but also with the specific concerns articulated in the Biden-Harris Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence. Aurora-M is rigorously evaluated across various tasks and languages, demonstrating robustness against catastrophic forgetting and outperforming alternatives in multilingual settings, particularly in safety evaluations. To promote responsible open-source LLM development, Aurora-M and its variants are released at https://huggingface.co/collections/aurora-m/aurora-m-models-65fdfdff62471e09812f5407 ., Comment: Preprint

Geological maps are an extremely valuable source of information for the Earth sciences. They provide insights into mineral exploration, vulnerability to natural hazards, and many other applications. These maps are created using numerical or conceptual models that use geological observations to extrapolate data. Geostatistical techniques have traditionally been used to generate reliable predictions that take into account the spatial patterns inherent in the data. However, as the number of auxiliary variables increases, these methods become more labor-intensive. Additionally, traditional machine learning methods often struggle with spatially correlated data and extracting valuable non-linear information from geoscientific datasets. To address these limitations, a new architecture called the Spatially Constrained Bayesian Network (SCB-Net) has been developed. The SCB-Net aims to effectively exploit the information from auxiliary variables while producing spatially constrained predictions. It is made up of two parts, the first part focuses on learning underlying patterns in the auxiliary variables while the second part integrates ground-truth data and the learned embeddings from the first part. Moreover, to assess model uncertainty, a technique called Monte Carlo dropout is used as a Bayesian approximation. The SCB-Net has been applied to two selected areas in northern Quebec, Canada, and has demonstrated its potential in generating field-data-constrained lithological maps while allowing assessment of prediction uncertainty for decision-making. This study highlights the promising advancements of deep neural networks in geostatistics, particularly in handling complex spatial feature learning tasks, leading to improved spatial information techniques., Comment: 17 pages, 3559 words, 14 figures

We construct a quantum critical Otto engine that is powered by finite temperature baths. We show that the work output of the engine shows universal power law behavior that depends on the critical exponents of the working medium, as well as on the temperature of the cold bath. Furthermore, higher temperatures of the cold bath allows the engine to approach the limit of adiabatic operation for smaller values of the time period, while the corresponding power shows a maximum at an intermediate value of the cold bath temperature. These counterintuitive results stems from thermal excitations dominating the dynamics at higher temperatures., Comment: 9 pages, 5 figures

Robust radiative transfer techniques are requisite for efficiently extracting the physical and chemical information from molecular rotational lines.We study several hypotheses that enable robust estimations of the column densities and physical conditions when fitting one or two transitions per molecular species. We study the extent to which simplifying assumptions aimed at reducing the complexity of the problem introduce estimation biases and how to detect them.We focus on the CO and HCO+ isotopologues and analyze maps of a 50 square arcminutes field. We used the RADEX escape probability model to solve the statistical equilibrium equations and compute the emerging line profiles, assuming that all species coexist. Depending on the considered set of species, we also fixed the abundance ratio between some species and explored different values. We proposed a maximum likelihood estimator to infer the physical conditions and considered the effect of both the thermal noise and calibration uncertainty. We analyzed any potential biases induced by model misspecifications by comparing the results on the actual data for several sets of species and confirmed with Monte Carlo simulations. The variance of the estimations and the efficiency of the estimator were studied based on the Cram{\'e}r-Rao lower bound.Column densities can be estimated with 30% accuracy, while the best estimations of the volume density are found to be within a factor of two. Under the chosen model framework, the peak 12CO(1--0) is useful for constraining the kinetic temperature. The thermal pressure is better and more robustly estimated than the volume density and kinetic temperature separately. Analyzing CO and HCO+ isotopologues and fitting the full line profile are recommended practices with respect to detecting possible biases.Combining a non-local thermodynamic equilibrium model with a rigorous analysis of the accuracy allows us to obtain an efficient estimator and identify where the model is misspecified. We note that other combinations of molecular lines could be studied in the future., Comment: Astronomy and Astrophysics - A\&A, In press

To address the global issue of online hate, hate speech detection (HSD) systems are typically developed on datasets from the United States, thereby failing to generalize to English dialects from the Majority World. Furthermore, HSD models are often evaluated on non-representative samples, raising concerns about overestimating model performance in real-world settings. In this work, we introduce NaijaHate, the first dataset annotated for HSD which contains a representative sample of Nigerian tweets. We demonstrate that HSD evaluated on biased datasets traditionally used in the literature consistently overestimates real-world performance by at least two-fold. We then propose NaijaXLM-T, a pretrained model tailored to the Nigerian Twitter context, and establish the key role played by domain-adaptive pretraining and finetuning in maximizing HSD performance. Finally, owing to the modest performance of HSD systems in real-world conditions, we find that content moderators would need to review about ten thousand Nigerian tweets flagged as hateful daily to moderate 60% of all hateful content, highlighting the challenges of moderating hate speech at scale as social media usage continues to grow globally. Taken together, these results pave the way towards robust HSD systems and a better protection of social media users from hateful content in low-resource settings., Comment: ACL 2024 main conference. Data and models available at https://github.com/worldbank/NaijaHate

Gallium phosphide (GaP) has recently received considerable attention as a suitable material for building photonic integrated circuits due to its remarkable optical and piezoelectric properties. Usually, GaP is grown epitaxially on III-V substrates to keep its crystallinity and later transferred to silicon wafers for further processing. Here, an alternative promising route for the fabrication of optomechanical (OM) cavities on GaP epitaxially grown on nominally (001)-oriented Si is introduced by using a two-step process consisting of a low-temperature etching of GaP followed by selective etching of the underneath silicon. The low-temperature (-30 $^o$C) during the dry-etching of GaP hinders the lateral etching rate, preserving the pattern with a deviation between the design and the pattern in the GaP layer lower than 5 %, avoiding the complex process of transferring and bonding a GaP wafer to a silicon-on-insulator wafer. To demonstrate the quality and feasibility of the proposed fabrication route, suspended OM cavities are fabricated and experimentally characterized. The cavities show optical quality factors between 10$^3$ and 10$^4$, and localized mechanical resonances at frequencies around 3.1 GHz. Both optical and mechanical resonances are close to those previously reported on crystalline GaP structures. These results suggest a simple and low-cost way to build GaP-based photonic devices directly integrated on industry-standard Si(001) photonic wafers., Comment: 11 pages, 5 figures

In this paper, we introduce the concept of using passive arm structures with intrinsic impedance for robot-robot and human-robot collaborative carrying with quadruped robots. The concept is meant for a leader-follower task and takes a minimalist approach that focuses on exploiting the robots' payload capabilities and reducing energy consumption, without compromising the robot locomotion capabilities. We introduce a preliminary arm mechanical design and describe how to use its joint displacements to guide the robot's motion. To control the robot's locomotion, we propose a decentralized Model Predictive Controller that incorporates an approximation of the arm dynamics and the estimation of the external forces from the collaborative carrying. We validate the overall system experimentally by performing both robot-robot and human-robot collaborative carrying on a stair-like obstacle and on rough terrain., Comment: The paper has 8 pages and 9 figures

Due to the spin-orbit coupling, Dirac fermions, submerged in a thermal bath with finite macroscopic vorticity, exhibit a spin polarisation along the direction parallel to the vorticity vector $\boldsymbol{\Omega}$. Due to the symmetries of the Lagrangian for free massless Dirac particles, there are three independent and classically conserved currents corresponding to the vector, axial, and helical charges. The constitutive relations for the charge currents and the stress-energy tensor at thermal equilibrium, derived in the framework of quantum field theory at finite temperature, reveal vorticity-induced contributions that deviate from the perfect fluid form. In this paper, we consider the mode structure of the corresponding hydrodynamical theory and derive collective excitations associated with coherent fluctuations of all three charges. We show that the chirally imbalanced rotating fluid should possess non-reciprocal gapless waves that propagate with different velocities along and opposite to the vorticity vector. We also uncover a strictly unidirectional mode, which we call the Axial Vortical Wave, propagating in the background of the axial charge density. We point out an unexpected instability in the limit of degenerate matter and discuss possible solutions when helicity and axial charge non-conservation is taken into account., Comment: 41 pages, 6 figures, 1 table. Part 1

In this paper, we consider the effect of interactions on the local, average polarization of a quantum plasma of massless fermion particles characterized by vector, axial, and helical quantum numbers. Due to the helical and axial vortical effects, perturbations in the vector charge in a rotating plasma can lead to chiral and helical charge transfer along the direction of the vorticity vector. At the same time, interactions between the plasma constituents lead to the dissipation of the helical charge through helicity-violating pair annihilation (HVPA) processes and of the axial charge through the axial anomaly. We will discuss separately a QED-like plasma, in which we ignore background electromagnetic fields and thus the axial charge is roughly conserved, as well as a QCD-like plasma, where instanton effects lead to the violation of the axial charge conservation, even in the absence of background chromomagnetic fields. The non-conservation of helicity and chirality leads to a gapping of the Helical, Axial and mixed, Axial-Helical vortical waves that prevents their infrared modes from propagating. On the other hand, usual dissipative effects, such as charge diffusion, lead to significant damping of ultraviolet modes. We end this paper with a discussion of the regimes where these vortical waves may propagate., Comment: 56 pages, 14 figures. Part 2

Well-designed current control is a key factor in ensuring the efficient and safe operation of modular multilevel converters (MMCs). Even though this control problem involves multiple control objectives, conventional current control schemes are comprised of independently designed decoupled controllers, e.g., proportional-integral (PI) or proportional-resonant (PR). Due to the bilinearity of the MMC dynamics, tuning PI and PR controllers so that good performance and constraint satisfaction are guaranteed is quite challenging. This challenge becomes more relevant in an AC/AC MMC configuration due to the complexity of tracking the single-phase sinusoidal components of the MMC output. In this paper, we propose a method to design a multivariable controller, i.e., a static feedback gain, to regulate the MMC currents. We use a physics-informed transformation to model the MMC dynamics linearly and synthesise the proposed controller. We use this linear model to formulate a linear matrix inequality that computes a feedback gain that guarantees safe and effective operation, including (i) limited tracking error, (ii) stability, and (iii) meeting all constraints. To test the efficacy of our method, we examine its performance in a direct AC/AC MMC simulated in Simulink/PLECS and in a scaled-down AC/AC MMC prototype to investigate the ultra-fast charging of electric vehicles., Comment: Submitted to IEEE Open Journal of the Industrial Electronics

Toxicity is a roadblock that prevents an inordinate number of drugs from being used in potentially life-saving applications. Deep learning provides a promising solution to finding ideal drug candidates; however, the vastness of chemical space coupled with the underlying $\mathcal{O}(n^3)$ matrix multiplication means these efforts quickly become computationally demanding. To remedy this, we present a hybrid quantum-classical neural network for predicting drug toxicity, utilizing a quantum circuit design that mimics classical neural behavior by explicitly calculating matrix products with complexity $\mathcal{O}(n^2)$. Leveraging the Hadamard test for efficient inner product estimation rather than the conventionally used swap test, we reduce the number qubits by half and remove the need for quantum phase estimation. Directly computing matrix products quantum mechanically allows for learnable weights to be transferred from a quantum to a classical device for further training. We apply our framework to the Tox21 dataset and show that it achieves commensurate predictive accuracy to the model's fully classical $\mathcal{O}(n^3)$ analog. Additionally, we demonstrate the model continues to learn, without disruption, once transferred to a fully classical architecture. We believe combining the quantum advantage of reduced complexity and the classical advantage of noise-free calculation will pave the way to more scalable machine learning models., Comment: This work is available as open source at https://github.com/anthonysmaldone/Quantum-to-Classical-Transfer-Learning

In this paper, we consider the inverse scattering problem associated with an anisotropic medium with a conductive boundary. We will assume that the corresponding far-field pattern is known/measured and we consider two inverse problems. First, we show that the far-field data uniquely determines the boundary coefficient. Next, since it is known that anisotropic coefficients are not uniquely determined by this data we will develop a qualitative method to recover the scatterer. To this end, we study the so-called monotonicity method applied to this inverse shape problem. This method has recently been applied to some inverse scattering problems but this is the first time it has been applied to an anisotropic scatterer. This method allows one to recover the scatterer but considering the eigenvalues of an operator associated with the far--field operator. We present some simple numerical reconstructions to illustrate our theory in two dimensions. For our reconstructions, we need to compute the adjoint of the Herglotz wave function as an operator mapping into $H^1$ of a small ball.

The Amazon Basin and the Latin America and Caribbean (LAC) region stands at a critical juncture, grappling with pressing environmental challenges while holding immense potential for transformative change through innovative solutions. This report illuminates the diverse landscape of social-ecological issues, technological advancements, community-led initiatives, and strategic actions that could help foster biosphere-based sustainability and resilience across the region., Comment: 38 pages, 11 figures

Companies with datacenters are procuring significant amounts of renewable energy to reduce their carbon footprint. There is increasing interest in achieving 24/7 Carbon-Free Energy (CFE) matching in electricity usage, aiming to eliminate all carbon footprints associated with electricity consumption on an hourly basis. However, the variability of renewable energy resources poses significant challenges for achieving this goal. We explore the impact of shifting computing jobs and associated power loads both in time and between datacenter locations. We develop an optimization model to simulate a network of geographically distributed datacenters managed by a company leveraging spatio-temporal load flexibility to achieve 24/7 CFE matching. We isolate three signals relevant for informed use of load flexiblity: varying average quality of renewable energy resources, low correlation between wind power generation over long distances due to different weather conditions, and lags in solar radiation peak due to Earth's rotation. We illustrate that the location of datacenters and the time of year affect which signal drives an effective load-shaping strategy. The energy procurement and load-shifting decisions based on informed use of these signals facilitate the resource-efficiency and cost-effectiveness of clean computing -- the costs of 24/7 CFE are reduced by 1.29$\pm$0.07 EUR/MWh for every additional percentage of flexible load. We provide practical guidelines on how companies with datacenters can leverage spatio-temporal load flexibility for truly clean computing. Our results and the open-source optimization model can also be useful for a broader variety of companies with flexible loads and an interest in eliminating their carbon footprint.

Twenty five years ago, several authors proposed to describe the forward interest rate curve (FRC) as an elastic string along which idiosyncratic shocks propagate, accounting for the peculiar structure of the return correlation across different maturities. In this paper, we revisit the specific "stiff'' elastic string field theory of Baaquie and Bouchaud (2004) in a way that makes its micro-foundation more transparent. Our model can be interpreted as capturing the effect of market forces that set the rates of nearby tenors in a self-referential fashion. The model is parsimonious and accurately reproduces the whole correlation structure of the FRC over the time period 1994-2023, with an error around 1% and with only one adjustable parameter, the value of which being very stable across the last three decades. The dependence of correlation on time resolution (also called the Epps effect) is also faithfully reproduced within the model and leads to a cross-tenor information propagation time on the order of 30 minutes. Finally, we confirm that the perceived time in interest rate markets is a strongly sub-linear function of real time, as surmised by Baaquie and Bouchaud (2004). In fact, our results are fully compatible with hyperbolic discounting, in line with the recent behavioral Finance literature (Farmer and Geanakoplos, 2009)., Comment: 21 pages, 12 figures

The Josephus problem is a well--studied elimination problem consisting in determining the position of the survivor after repeated applications of a deterministic rule removing one person at a time from a given group. A natural probabilistic variant of this process is introduced in this paper. More precisely, in this variant, the survivor is determined after performing a succession of Bernouilli trials with parameter $p$ designating each time the person to remove. When the number of participants tends to infinity, the main result characterises the limit distribution of the position of the survivor with an increasing degree of precision as the parameter approaches the unbiaised case $p=1/2$. Then, the convergence rate to the position of the survivor is obtained in the form of a Central-Limit Theorem. A number of other variants of the suggested probabilistic elimination process are also considered. They each admit a specific limit behavior which, in most cases, is stated in the form of an open problem.

Recent techniques for text-to-4D generation synthesize dynamic 3D scenes using supervision from pre-trained text-to-video models. However, existing representations for motion, such as deformation models or time-dependent neural representations, are limited in the amount of motion they can generate-they cannot synthesize motion extending far beyond the bounding box used for volume rendering. The lack of a more flexible motion model contributes to the gap in realism between 4D generation methods and recent, near-photorealistic video generation models. Here, we propose TC4D: trajectory-conditioned text-to-4D generation, which factors motion into global and local components. We represent the global motion of a scene's bounding box using rigid transformation along a trajectory parameterized by a spline. We learn local deformations that conform to the global trajectory using supervision from a text-to-video model. Our approach enables the synthesis of scenes animated along arbitrary trajectories, compositional scene generation, and significant improvements to the realism and amount of generated motion, which we evaluate qualitatively and through a user study. Video results can be viewed on our website: https://sherwinbahmani.github.io/tc4d., Comment: Project Page: https://sherwinbahmani.github.io/tc4d

This article considers a semi-supervised classification setting on a Gaussian mixture model, where the data is not labeled strictly as usual, but instead with uncertain labels. Our main aim is to compute the Bayes risk for this model. We compare the behavior of the Bayes risk and the best known algorithm for this model. This comparison eventually gives new insights over the algorithm.

We consider a discrete-time system of n coupled random vectors, a.k.a. interacting particles. The dynamics involve a vanishing step size, some random centered perturbations, and a mean vector field which induces the coupling between the particles. We study the doubly asymptotic regime where both the number of iterations and the number n of particles tend to infinity, without any constraint on the relative rates of convergence of these two parameters. We establish that the empirical measure of the interpolated trajectories of the particles converges in probability, in an ergodic sense, to the set of recurrent Mc-Kean-Vlasov distributions. A first application example is the granular media equation, where the particles are shown to converge to a critical point of the Helmholtz energy. A second example is the convergence of stochastic gradient descent to the global minimizer of the risk, in a wide two-layer neural networks using random features.

Safe path and gait planning are essential for bipedal robots to navigate complex real-world environments. The prevailing approaches often plan the path and gait separately in a hierarchical fashion, potentially resulting in unsafe movements due to neglecting the physical constraints of walking robots. A safety-critical path must not only avoid obstacles but also ensure that the robot's gaits are subject to its dynamic and kinematic constraints. This work presents a novel approach that unifies path planning and gait planning via a Model Predictive Control (MPC) using the Linear Inverted Pendulum (LIP) model representing bipedal locomotion. This approach considers environmental constraints, such as obstacles, and the robot's kinematics and dynamics constraints. By using discrete-time Control Barrier Functions for obstacle avoidance, our approach generates the next foot landing position, ensuring robust walking gaits and a safe navigation path within clustered environments. We validated our proposed approach in simulation using a Digit robot in 20 randomly created environments. The results demonstrate improved performance in terms of safety and robustness when compared to hierarchical path and gait planning frameworks., Comment: 8 pages

Traditional one-step preview planning algorithms for bipedal locomotion struggle to generate viable gaits when walking across terrains with restricted footholds, such as stepping stones. To overcome such limitations, this paper introduces a novel multi-step preview foot placement planning algorithm based on the step-to-step discrete evolution of the Divergent Component of Motion (DCM) of walking robots. Our proposed approach adaptively changes the step duration and the swing foot trajectory for optimal foot placement under constraints, thereby enhancing the long-term stability of the robot and significantly improving its ability to navigate environments with tight constraints on viable footholds. We demonstrate its effectiveness through various simulation scenarios with complex stepping-stone configurations and external perturbations. These tests underscore its improved performance for navigating foothold-restricted terrains, even with external disturbances., Comment: 7 pages, 7 figures, submitted to ICRA 2025, for associated simulation video, see https://youtu.be/DjH69m1kbnM

Efficient path planning for autonomous mobile robots is a critical problem across numerous domains, where optimizing both time and energy consumption is paramount. This paper introduces a novel methodology that considers the dynamic influence of an environmental flow field and considers geometric constraints, including obstacles and forbidden zones, enriching the complexity of the planning problem. We formulate it as a multi-objective optimal control problem, propose a novel transformation called Harmonic Transformation, and apply a semi-Lagrangian scheme to solve it. The set of Pareto efficient solutions is obtained considering two distinct approaches: a deterministic method and an evolutionary-based one, both of which are designed to make use of the proposed Harmonic Transformation. Through an extensive analysis of these approaches, we demonstrate their efficacy in finding optimized paths., Comment: 12 pages, initial paper submission; Preprint submitted to Journal of the Franklin Institute

We introduce quantum circuits for simulations of multi-mode state-vectors on 3D cQED processors, using matrix product state representations. The circuits are demonstrated as applied to simulations of molecular docking based on holographic Gaussian boson sampling, as illustrated for binding of a thiol-containing aryl sulfonamide ligand to the tumor necrosis factor-$\alpha$ converting enzyme receptor. We show that cQED devices with a modest number of modes could be employed to simulate multimode systems by re-purposing working modes through measurement and re-initialization. We anticipate a wide range of GBS applications could be implemented on compact 3D cQED processors analogously, using the holographic approach. Simulations on qubit-based quantum computers could be implemented analogously, using circuits that represent continuous variables in terms of truncated expansions of Fock states.

Nano-drones, distinguished by their agility, minimal weight, and cost-effectiveness, are particularly well-suited for exploration in confined, cluttered and narrow spaces. Recognizing transparent, highly reflective or absorbing materials, such as glass and metallic surfaces is challenging, as classical sensors, such as cameras or laser rangers, often do not detect them. Inspired by bats, which can fly at high speeds in complete darkness with the help of ultrasound, this paper introduces \textit{BatDeck}, a pioneering sensor-deck employing a lightweight and low-power ultrasonic sensor for nano-drone autonomous navigation. This paper first provides insights about sensor characteristics, highlighting the influence of motor noise on the ultrasound readings, then it introduces the results of extensive experimental tests for obstacle avoidance (OA) in a diverse environment. Results show that \textit{BatDeck} allows exploration for a flight time of 8 minutes while covering 136m on average before crash in a challenging environment with transparent and reflective obstacles, proving the effectiveness of ultrasonic sensors for OA on nano-drones., Comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

The flux of atmospheric and astrophysical neutrinos measured in UHE neutrino detectors is strongly dependent on the description of the propagation and absorption of the neutrinos during the passage through Earth to the detector. In particular, the attenuation of the incident neutrino flux depends on the details of the matter structure between the source and the detector. In this paper, we will investigate the impact of different descriptions for the density profile of Earth on the transmission coefficient, defined as the ratio between the flux measured in the detector and the incoming neutrino flux. We will consider five different models for the Earth's density profile and estimate how these different models modify the target column density and transmission coefficients for different flavors. The results are derived by solving the cascade equations taking into account the NC interactions and tau regeneration. A comparison with approximated solutions is also presented. Our results indicated that the predictions are sensitive to the model considered for the density profile, with the simplified three layer model providing a satisfactory description when compared with the PREM results., Comment: 10 pages, 6 figure, 1 table

Recent developments in unmanned aerial vehicles (UAVs) and mobile edge computing (MEC) have provided users with flexible and resilient computing services. However, meeting the computing-intensive and latency-sensitive demands of users poses a significant challenge due to the limited resources of UAVs. To address this challenge, we present a multi-objective optimization approach for multi-UAV-assisted MEC systems. First, we formulate a multi-objective optimization problem \textcolor{b2}{aiming} at minimizing the total task completion delay, reducing the total UAV energy consumption, and maximizing the total amount of offloaded tasks by jointly optimizing task offloading, computation resource allocation, and UAV trajectory control. Since the problem is a mixed-integer non-linear programming (MINLP) and NP-hard problem which is challenging, we propose a joint task offloading, computation resource allocation, and UAV trajectory control (JTORATC) approach to solve the problem. \textcolor{b3}{However, since the decision variables of task offloading, computation resource allocation, and UAV trajectory control are coupled with each other, the original problem is split into three sub-problems, i.e., task offloading, computation resource allocation, and UAV trajectory control, which are solved individually to obtain the corresponding decisions.} \textcolor{b2}{Moreover, the sub-problem of task offloading is solved by using distributed splitting and threshold rounding methods, the sub-problem of computation resource allocation is solved by adopting the Karush-Kuhn-Tucker (KKT) method, and the sub-problem of UAV trajectory control is solved by employing the successive convex approximation (SCA) method.} Simulation results show that the proposed JTORATC has superior performance compared to the other benchmark methods.

We consider discrete Schr\"odinger operators on $\ell^2(\mathbb{Z})$ with bounded random but not necessarily identically distributed values of the potential. We prove spectral localization (with exponentially decaying eigenfunctions) as well as dynamical localization for this model. An important ingredient of the proof is a non-stationary version of the parametric Furstenberg Theorem on random matrix products, which is also of independent interest., Comment: 62 pages. arXiv admin note: text overlap with arXiv:1809.00416

Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) is emerging as a promising paradigm to provide aerial-terrestrial computing services in close proximity to mobile devices (MDs). However, meeting the demands of computation-intensive and delay-sensitive tasks for MDs poses several challenges, including the demand-supply contradiction between MDs and MEC servers, the demand-supply heterogeneity between MDs and MEC servers, the trajectory control requirements on energy efficiency and timeliness, and the different time-scale dynamics of the network. To address these issues, we first present a hierarchical architecture by incorporating terrestrial-aerial computing capabilities and leveraging UAV flexibility. Furthermore, we formulate a joint computing resource allocation, computation offloading, and trajectory control problem to maximize the system utility. Since the problem is a non-convex and NP-hard mixed integer nonlinear programming (MINLP), we propose a two-timescale joint computing resource allocation, computation offloading, and trajectory control (TJCCT) approach for solving the problem. In the short timescale, we propose a price-incentive model for on-demand computing resource allocation and a matching mechanism-based method for computation offloading. In the long timescale, we propose a convex optimization-based method for UAV trajectory control. Besides, we theoretically prove the stability, optimality, and polynomial complexity of TJCCT. Extended simulation results demonstrate that the proposed TJCCT outperforms the comparative algorithms in terms of the system utility, average processing rate, average completion delay, and average completion ratio.

In recent years, Deep Learning has gained popularity for its ability to solve complex classification tasks, increasingly delivering better results thanks to the development of more accurate models, the availability of huge volumes of data and the improved computational capabilities of modern computers. However, these improvements in performance also bring efficiency problems, related to the storage of datasets and models, and to the waste of energy and time involved in both the training and inference processes. In this context, data reduction can help reduce energy consumption when training a deep learning model. In this paper, we present up to eight different methods to reduce the size of a tabular training dataset, and we develop a Python package to apply them. We also introduce a representativeness metric based on topology to measure how similar are the reduced datasets and the full training dataset. Additionally, we develop a methodology to apply these data reduction methods to image datasets for object detection tasks. Finally, we experimentally compare how these data reduction methods affect the representativeness of the reduced dataset, the energy consumption and the predictive performance of the model.

Graph convolutional networks (GCNs) have emerged as a powerful alternative to multiple instance learning with convolutional neural networks in digital pathology, offering superior handling of structural information across various spatial ranges - a crucial aspect of learning from gigapixel H&E-stained whole slide images (WSI). However, graph message-passing algorithms often suffer from oversmoothing when aggregating a large neighborhood. Hence, effective modeling of multi-range interactions relies on the careful construction of the graph. Our proposed multi-scale GCN (MS-GCN) tackles this issue by leveraging information across multiple magnification levels in WSIs. MS-GCN enables the simultaneous modeling of long-range structural dependencies at lower magnifications and high-resolution cellular details at higher magnifications, akin to analysis pipelines usually conducted by pathologists. The architecture's unique configuration allows for the concurrent modeling of structural patterns at lower magnifications and detailed cellular features at higher ones, while also quantifying the contribution of each magnification level to the prediction. Through testing on different datasets, MS-GCN demonstrates superior performance over existing single-magnification GCN methods. The enhancement in performance and interpretability afforded by our method holds promise for advancing computational pathology models, especially in tasks requiring extensive spatial context.

Considering two antipodal observers in de Sitter space, we illustrate how spacetime connectivity between the holographic screens located on the (stretched) horizons emerges from holographic entanglement. To do so, we construct a covariant holographic entanglement entropy prescription in de Sitter space, including quantum corrections. While this prescription is inspired by the bilayer proposal, we argue that the monolayer proposal appears to be inconsistent. Entanglement wedge reconstruction implies an extension of static patch holography where the exterior region connecting the static patches is included, and reconstructible from the two screens. A phase transition occurs where there is an exchange of dominance between two competing quantum extremal surfaces, leading to a transfer of the encoding of the exterior region from one screen to the other. The effects in the bulk of integrating out degrees of freedom from the screens are discussed., Comment: Based on a talk given at the Corfu Summer Institute 2023 "School and Workshops on Elementary Particle Physics and Gravity'' (CORFU2023), 23 April - 6 May , and 27 August - 1 October, 2023, Corfu, Greece. 16 pages + references, 5 Figures

The aim of this paper is to evaluate whether large language models trained on multi-choice question data can be used to discriminate between medical subjects. This is an important and challenging task for automatic question answering. To achieve this goal, we train deep neural networks for multi-class classification of questions into the inferred medical subjects. Using our Multi-Question (MQ) Sequence-BERT method, we outperform the state-of-the-art results on the MedMCQA dataset with an accuracy of 0.68 and 0.60 on their development and test sets, respectively. In this sense, we show the capability of AI and LLMs in particular for multi-classification tasks in the Healthcare domain.

Scandium aluminum nitride (ScAlN) has recently emerged as an attractive material for integrated photonics due to its favorable nonlinear optical properties and compatibility with CMOS fabrication. Despite the promising and versatile material properties, it is still an outstanding challenge to realize low-loss photonic circuits on thin-film ScAlN-on-insulator wafers. Here, we present a systematic study on the material quality of sputtered thin-film ScAlN produced in a CMOS-compatible 200 mm line, and an optimized fabrication process to yield 400 nm thick, fully etched waveguides. With surface polishing and annealing, we achieve micro-ring resonators with an intrinsic quality factor as high as $1.47\times 10^5$, corresponding to a propagation loss of 2.4 dB/cm. These results serve as a critical step towards developing future large-scale, low-loss photonic integrated circuits based on ScAlN.

The action of an ideal fluid in Euler variables with a variable number of particles is used for the phenomenological description of the processes of particle creation in strong external fields. It has been demonstrated that the conformal invariance of the creation law imposes quite strict restrictions on the possible types of sources. It is shown that combinations with the particle number density in the creation law can be interpreted as dark matter within the framework of this model., Comment: This article belongs to the Special Issue The Friedmann Cosmology: A Century Later

The ever increasing demand for computational power combined with the predicted plateau for the miniaturization of existing silicon-based technologies has made the search for low power alternatives an industrial and scientifically engaging problem. In this work, we explore spintronics-based Ising machines as hardware computation accelerators. We start by presenting the physical platforms on which this emerging field is being developed, the different control schemes and the type of algorithms and problems on which these machines outperform conventional computers. We then benchmark these technologies and provide an outlook for future developments and use-cases that can help them get a running start for integration into the next generation of computing devices.

This paper presents MULAN-WC, a novel multi-robot 3D reconstruction framework that leverages wireless signal-based coordination between robots and Neural Radiance Fields (NeRF). Our approach addresses key challenges in multi-robot 3D reconstruction, including inter-robot pose estimation, localization uncertainty quantification, and active best-next-view selection. We introduce a method for using wireless Angle-of-Arrival (AoA) and ranging measurements to estimate relative poses between robots, as well as quantifying and incorporating the uncertainty embedded in the wireless localization of these pose estimates into the NeRF training loss to mitigate the impact of inaccurate camera poses. Furthermore, we propose an active view selection approach that accounts for robot pose uncertainty when determining the next-best views to improve the 3D reconstruction, enabling faster convergence through intelligent view selection. Extensive experiments on both synthetic and real-world datasets demonstrate the effectiveness of our framework in theory and in practice. Leveraging wireless coordination and localization uncertainty-aware training, MULAN-WC can achieve high-quality 3d reconstruction which is close to applying the ground truth camera poses. Furthermore, the quantification of the information gain from a novel view enables consistent rendering quality improvement with incrementally captured images by commending the robot the novel view position. Our hardware experiments showcase the practicality of deploying MULAN-WC to real robotic systems.