Search

Causal discovery is essential for understanding complex systems, yet traditional methods often depend on strong, untestable assumptions, making the process challenging. Large Language Models (LLMs) present a promising alternative for extracting causal insights from text-based metadata, which consolidates domain expertise. However, LLMs are prone to unreliability and hallucinations, necessitating strategies that account for their limitations. One such strategy involves leveraging a consistency measure to evaluate reliability. Additionally, most text metadata does not clearly distinguish direct causal relationships from indirect ones, further complicating the inference of causal graphs. As a result, focusing on causal orderings, rather than causal graphs, emerges as a more practical and robust approach. We propose a novel method to derive a distribution of acyclic tournaments (representing plausible causal orders) that maximizes a consistency score. Our approach begins by computing pairwise consistency scores between variables, yielding a cyclic tournament that aggregates these scores. From this structure, we identify optimal acyclic tournaments compatible with the original tournament, prioritizing those that maximize consistency across all configurations. We tested our method on both classical and well-established bechmarks, as well as real-world datasets from epidemiology and public health. Our results demonstrate the effectiveness of our approach in recovering distributions causal orders with minimal error.

Singlet exciton fission has the potential to increase the efficiency of crystalline silicon solar cells beyond the conventional single junction limit. Perhaps the largest obstacle to achieving this enhancement is uncertainty about energy coupling mechanisms at the interfaces between silicon and exciton fission materials such as tetracene. Here, the previously reported silicon-hafnium oxynitride-tetracene structure is studied and a combination of magnetic-field-dependent silicon photoluminescence measurements and density functional theory calculations is used to probe the influence of the interlayer composition on the triplet transfer process across the hafnium oxynitride interlayer. It is found that hafnium oxide interlayers do not show triplet exciton sensitization of silicon, and that nitrogen content in hafnium oxynitride layers is correlated with enhanced sensitization. Calculation results reveal that defects in hafnium oxynitride interlayers with higher nitrogen content introduce states close to the band-edge of silicon, which can mediate the triplet exciton transfer process. Some defects introduce additional deleterious mid-gap states, which may explain observed silicon photoluminescence quenching. These results show that band-edge states can mediate the triplet exciton transfer process, potentially through a sequential charge transfer mechanism., Comment: 30 pages, 20 figures

In this work, we developed new mathematical methods for analyzing network topology and applied these methods to the analysis of brain networks. More specifically, we rigorously developed quantitative methods based on complexes constructed from digraphs (digraph-based complexes), such as path complexes and directed clique complexes (alternatively, we refer to these complexes as "higher-order structures," or "higher-order topologies," or "simplicial structures"), and, in the case of directed clique complexes, also methods based on the interrelations between the directed cliques, what we called "directed higher-order connectivities." This new quantitative theory for digraph-based complexes can be seen as a step towards the formalization of a "quantitative simplicial theory." Subsequently, we used these new methods, such as characterization measures and similarity measures for digraph-based complexes, to analyze the topology of digraphs derived from brain connectivity estimators, specifically the estimator known as information partial directed coherence (iPDC), which is a multivariate estimator that can be considered a representation of Granger causality in the frequency-domain, particularly estimated from electroencephalography (EEG) data from patients diagnosed with left temporal lobe epilepsy, in the delta, theta and alpha frequency bands, to try to find new biomarkers based on the higher-order structures and connectivities of these digraphs. In particular, we attempted to answer the following questions: How does the higher-order topology of the brain network change from the pre-ictal to the ictal phase, from the ictal to the post-ictal phase, at each frequency band and in each cerebral hemisphere? Does the analysis of higher-order structures provide new and better biomarkers for seizure dynamics and also for the laterality of the seizure focus than the usual graph theoretical analyses?

While silicon solar cells dominate global photovoltaic energy production, their continued improvement is hindered by the single junction limit. One potential solution is to use molecular singlet exciton fission to generate two electrons from each absorbed high-energy photon. We demonstrate that the long-standing challenge of coupling molecular excited states to silicon solar cells can be overcome using sequential charge transfer. Combining zinc phthalocyanine, aluminum oxide, and a shallow junction crystalline silicon microwire solar cell, the peak charge generation efficiency per photon absorbed in tetracene is (138 +- 6)%, comfortably surpassing the quantum efficiency limit for conventional silicon solar cells and establishing a new, scalable approach to low cost, high efficiency photovoltaics., Comment: 34 pages, 14 figures

Physical devices exhibiting stochastic functions with low energy consumption and high device density have the potential to enable complex probability-based computing algorithms, accelerate machine learning tasks, and enhance hardware security. Recently, superparamagnetic tunnel junctions (sMTJs) have been widely explored for such purposes, leading to the development of sMTJ-based systems; however, the reliance on nanoscale ferromagnets limits scalability and reliability, making sMTJs sensitive to external perturbations and prone to significant device variations. Here, we present an experimental demonstration of closed loop three-terminal sMTJs as reliable and potentially scalable sources of true randomness in the field-free regime. By leveraging dual-current controllability and incorporating feedback, we stabilize the switching operation of superparamagnets and reach cryptographic-quality random bitstreams. The realization of controllable and robust true random sMTJs underpin a general hardware platform for computing schemes exploiting the stochasticity in the physical world, as demonstrated by the generative artificial intelligence example in our experiment.

In this study, we explore how quantum pathfinding algorithm called Gaussian Amplitude Amplification (GAA) can be used to solve the DNA sequencing problem. To do this, sequencing by hybridization was assumed wherein short fragments of the nucleic acids called oligonucleotides of length l were gathered and were then assembled. The process of reassembling the sequence was then abstracted into a graph problem of finding the Hamiltonian path with the least cost. The constructed directed graph was then converted into sequential bipartite graphs in order to use GAA. The results of our simulation revealed that for the case where l = 2 and spectrum size of |S| = 4, the probability of finding the optimal solution (with the least cost) is approximately 70.92% - a significant improvement compared to 4.17% when the path is chosen randomly. While this study only focused on the ideal scenario where there are no errors in the spectrum, the outcomes presented here demonstrate the plausibility of using GAA as a genome sequencing method., Comment: 13 pages, 8 figures

To fight climate change, new "green" technology are emerging, most of them using electricity as a power source. Among the solutions, Electric Vehicles (EVs) represent a central asset in the future transport system. EVs require a complex infrastructure to enable the so-called Vehicle-to-Grid (V2G) paradigm to manage the charging process between the smart grid and the EV. In this paradigm, the Electric Vehicle Supply Equipment (EVSE), or charging station, is the end device that authenticates the vehicle and delivers the power to charge it. However, since an EVSE is publicly exposed and connected to the Internet, recent works show how an attacker with physical tampering and remote access can target an EVSE, exposing the security of the entire infrastructure and the final user. For this reason, it is important to develop novel strategies to secure such infrastructures. In this paper we present HoneyEVSE, the first honeypot conceived to simulate an EVSE. HoneyEVSE can simulate with high fidelity the EV charging process and, at the same time, enables a user to interact with it through a dashboard. Furthermore, based on other charging columns exposed on the Internet, we emulate the login and device information pages to increase user engagement. We exposed HoneyEVSE for 30 days to the Internet to assess its capability and measured the interaction received with its Shodan Honeyscore. Results show that HoneyEVSE can successfully evade the Shodan honeyscore metric while attracting a high number of interactions on the exposed services., Comment: 15 pages

Air Quality Monitoring and Forecasting has been a popular research topic in recent years. Recently, data-driven approaches for air quality forecasting have garnered significant attention, owing to the availability of well-established data collection facilities in urban areas. Fixed infrastructures, typically deployed by national institutes or tech giants, often fall short in meeting the requirements of diverse personalized scenarios, e.g., forecasting in areas without any existing infrastructure. Consequently, smaller institutes or companies with limited budgets are compelled to seek tailored solutions by introducing more flexible infrastructures for data collection. In this paper, we propose an expandable graph attention network (EGAT) model, which digests data collected from existing and newly-added infrastructures, with different spatial structures. Additionally, our proposal can be embedded into any air quality forecasting models, to apply to the scenarios with evolving spatial structures. The proposal is validated over real air quality data from PurpleAir., Comment: Accepted by DSAA'23 (Industrial Track)

Air quality forecasting has garnered significant attention recently, with data-driven models taking center stage due to advancements in machine learning and deep learning models. However, researchers face challenges with complex data acquisition and the lack of open-sourced datasets, hindering efficient model validation. This paper introduces PurpleAirSF, a comprehensive and easily accessible dataset collected from the PurpleAir network. With its high temporal resolution, various air quality measures, and diverse geographical coverage, this dataset serves as a useful tool for researchers aiming to develop novel forecasting models, study air pollution patterns, and investigate their impacts on health and the environment. We present a detailed account of the data collection and processing methods employed to build PurpleAirSF. Furthermore, we conduct preliminary experiments using both classic and modern spatio-temporal forecasting models, thereby establishing a benchmark for future air quality forecasting tasks., Comment: Accepted by ACM SIGSPATIAL 2023

In the last decade, the scientific community has devolved its attention to the deployment of data-driven approaches in scientific research to provide accurate and reliable analysis of a plethora of phenomena. Most notably, Physics-informed Neural Networks and, more recently, Universal Differential Equations (UDEs) proved to be effective both in system integration and identification. However, there is a lack of an in-depth analysis of the proposed techniques. In this work, we make a contribution by testing the UDE framework in the context of Ordinary Differential Equations (ODEs) discovery. In our analysis, performed on two case studies, we highlight some of the issues arising when combining data-driven approaches and numerical solvers, and we investigate the importance of the data collection process. We believe that our analysis represents a significant contribution in investigating the capabilities and limitations of Physics-informed Machine Learning frameworks.

We investigate the effects of single, multiple, and extended defects in the form of non-magnetic impurities and vacancies in twisted bilayer graphene (TBG) at and away from the magic angle, using a fully atomistic model and focusing on the behavior of the flat low-energy moir\'e bands. For strong impurities and vacancies in the $AA$ region we find a complete removal of one of the four moir\'e bands, resulting in a significant depletion of the charge density in the $AA$ regions even at extremely low defect concentrations. We find similar results for other defect locations, with the exception of the least coordinated sites in the $AB$ region, where defects instead result in a peculiar band replacement process within the moir\'e bands. In the vacancy limit, this process yields a band structure misleadingly similar to the pristine case. Moreover, we show that triple point fermions (TPFs), which are the crossing of the Dirac point by a flat band, appearing for single, periodic, defects, are generally not preserved when adding extended or multiple defects, and thus likely not experimentally relevant. We further identify two universal length scales for defects, consisting of charge modulations on the atomic scale and on the moir\'e scale, illustrating the importance of both the atomic and moir\'e structures for understanding TBG. We show that our conclusions hold beyond the magic angle and for fully isolated defects. In summary, our results demonstrate that the normal state of TBG and its moir\'e flat bands are extremely sensitive to both the location and strength of non-magnetic impurities and vacancies, which should have significant implications for any emergent ordered state., Comment: 15 pages, 10 figures

Denoising Diffusion models have shown remarkable performance in generating diverse, high quality images from text. Numerous techniques have been proposed on top of or in alignment with models like Stable Diffusion and Imagen that generate images directly from text. A lesser explored approach is DALLE-2's two step process comprising a Diffusion Prior that generates a CLIP image embedding from text and a Diffusion Decoder that generates an image from a CLIP image embedding. We explore the capabilities of the Diffusion Prior and the advantages of an intermediate CLIP representation. We observe that Diffusion Prior can be used in a memory and compute efficient way to constrain the generation to a specific domain without altering the larger Diffusion Decoder. Moreover, we show that the Diffusion Prior can be trained with additional conditional information such as color histogram to further control the generation. We show quantitatively and qualitatively that the proposed approaches perform better than prompt engineering for domain specific generation and existing baselines for color conditioned generation. We believe that our observations and results will instigate further research into the diffusion prior and uncover more of its capabilities.

We propose a method to distinguish between trivial and topological, Majorana, zero-energy states in both short and long superconductor-normal-superconductor junctions based on Rashba nanowires using phase-biased equilibrium transport measurements. In particular, we show how the sawtooth profile of the supercurrent, due to the Majorana oscillation suppression in the topological phase for sufficiently long superconductor regions, leads to a strong signal in its zero-frequency susceptibility for a phase difference of $\phi=\pi$. This signal is notably insensitive to the chemical potential in the normal region, while trivial zero-energy states only causes signals in the susceptibility that is highly varying with the chemical potential, thus turning gating of the normal region into a simple experimental control knob. Furthermore, we obtain that, by tuning the junction transparency, critical currents in both short and long junctions undergo a reduction in the number of oscillations as a function of magnetic field only in the topological phase, an effect that find to be intimately linked to Majorana non-locality. Finally, we show that our results also hold at finite temperatures, thus highlighting their potential measurability under realistic experimental conditions., Comment: 12 pages, 7 figures. Published version

We present HyperNST; a neural style transfer (NST) technique for the artistic stylization of images, based on Hyper-networks and the StyleGAN2 architecture. Our contribution is a novel method for inducing style transfer parameterized by a metric space, pre-trained for style-based visual search (SBVS). We show for the first time that such space may be used to drive NST, enabling the application and interpolation of styles from an SBVS system. The technical contribution is a hyper-network that predicts weight updates to a StyleGAN2 pre-trained over a diverse gamut of artistic content (portraits), tailoring the style parameterization on a per-region basis using a semantic map of the facial regions. We show HyperNST to exceed state of the art in content preservation for our stylized content while retaining good style transfer performance.

We present StyleBabel, a unique open access dataset of natural language captions and free-form tags describing the artistic style of over 135K digital artworks, collected via a novel participatory method from experts studying at specialist art and design schools. StyleBabel was collected via an iterative method, inspired by `Grounded Theory': a qualitative approach that enables annotation while co-evolving a shared language for fine-grained artistic style attribute description. We demonstrate several downstream tasks for StyleBabel, adapting the recent ALADIN architecture for fine-grained style similarity, to train cross-modal embeddings for: 1) free-form tag generation; 2) natural language description of artistic style; 3) fine-grained text search of style. To do so, we extend ALADIN with recent advances in Visual Transformer (ViT) and cross-modal representation learning, achieving a state of the art accuracy in fine-grained style retrieval.