Search

In today's data-driven era, computational systems generate vast amounts of data that drive the digital transformation of industries, where Artificial Intelligence (AI) plays a key role. Currently, the demand for eXplainable AI (XAI) has increased to enhance the interpretability, transparency, and trustworthiness of AI models. However, evaluating XAI methods remains challenging: existing evaluation frameworks typically focus on quantitative properties such as fidelity, consistency, and stability without taking into account qualitative characteristics such as satisfaction and interpretability. In addition, practitioners face a lack of guidance in selecting appropriate datasets, AI models, and XAI methods -a major hurdle in human-AI collaboration. To address these gaps, we propose a framework that integrates quantitative benchmarking with qualitative user assessments through virtual personas based on the "Anthology" of backstories of the Large Language Model (LLM). Our framework also incorporates a content-based recommender system that leverages dataset-specific characteristics to match new input data with a repository of benchmarked datasets. This yields an estimated XAI score and provides tailored recommendations for both the optimal AI model and the XAI method for a given scenario., Comment: 8 pages, 6 figures

Biological evolution continuously refines the design of species, resulting in highly optimised organisms over hundreds of millennia. Intuitively, we expect that random changes-evolution's primary mechanism-are more likely to be harmful than beneficial, leading to widespread detrimental effects in evolving species. The Evolvable Soma Theory of Ageing (ESTA) suggests that ageing is the cumulative result of these harmful effects, which predominantly cause bodily damage, while a few may lead to beneficial adaptations that evolution can exploit. While the disposable soma theory views ageing as a consequence of limited evolutionary pressure, ESTA posits that ageing is essentially evolution in action. In this study, we gather evidence supporting this theory through computer simulations. We conduct experiments using a platform where genes are linked to onset values that determine when they are expressed. Three scenarios are tested: one with single-point fitness evaluation, constant mutation rate and fixed gene onsets; one with single-point fitness evaluation, onset-dependent mutation rate and fixed gene onsets; and one with spread fitness evaluation, onset-dependent mutation rate and evolvable gene onsets. The last scenario, which embodies the evolvable soma hypothesis, demonstrates superior performance in both algorithmic efficiency and biological plausibility compared to the others.

Data exchange through mobile devices is rapidly increasing due to the high information demands of today's applications. The need for monitoring the exchanged traffic becomes important in order to control and optimize the device and network performance and security. Taking this under consideration, in this paper, we developed a cloud-based system for the analysis of network traffic. The smartphone devices act both as traffic captors and visualization endpoints, enabling the user to get an overview of the network while minimizing resource consumption. In the presented work, we evaluate our system using two test cases and a variety of target devices. Our results prove the usefulness of the proposed system architecture.

Besides major advantages for telecommunication applications, VCSELs have attracted interest for their potential for neuro-inspired computing, frequency comb generation or high-frequency spin oscillations. In the meantime, strain applied to the laser structure has been shown to have a significant impact on the laser emission properties such as the polarization dynamics or birefringence. In this work, we further explore the influence of strain on VCSELs and how this effect could be used to fine tune the laser wavelength. Through a comprehensive investigation, we demonstrate consistent wavelength shift up to 1 nm and report a sensitivity between 0.12 to 0.18 nm / millistrain. We also record birefringence values up to 292 GHz. Our results show that controlled strain level could be considered for fine wavelength tuning and possibly alleviate the selection of VCSEL for precise wavelength requirements., Comment: 4 pages, 4 figures, revised manuscript

We investigate the effect of a dark matter caustic passing through the Solar System. We find, confirming a previous result, that the Sun tracks the caustic surface for some time. We integrate numerically the equations of motion of the Sun and a comet for a large number of initial conditions and of caustic passage properties. We calculate the probability for the comet to escape the Solar System and the probability for it to fall within 50 A.U. of the Sun, given the initial semimajor axis and eccentricity of its orbit. We find that the average probability for a comet to fall within 50 A.U. of the Sun is of order $3 \times 10^{-4}$ and that comets which are initially at a distance larger than about $10^5$ A.U. have a probability of order one to be ejected from the Solar System.

At any moment in time, evolution is faced with a formidable challenge: refining the already highly optimised design of biological species, a feat accomplished through all preceding generations. In such a scenario, the impact of random changes (the method employed by evolution) is much more likely to be harmful than advantageous, potentially lowering the reproductive fitness of the affected individuals. Our hypothesis is that ageing is, at least in part, caused by the cumulative effect of all the experiments carried out by evolution to improve a species' design. These experiments are almost always unsuccessful, as expected given their pseudorandom nature, cause harm to the body and ultimately lead to death. On the other hand, a small minority of experiments have positive outcome, offering valuable insight into the direction evolution should pursue. This hypothesis is consistent with the concept of "terminal addition", by which nature is biased towards adding innovations at the end of development. From the perspective of evolution as an optimisation algorithm, ageing is advantageous as it allows to test innovations during a phase when their impact on fitness is present but less pronounced. Our inference suggests that ageing has a key biological role, as it contributes to the system's evolvability by exerting a regularisation effect on the fitness landscape of evolution.

The field of Explainable Artificial Intelligence (XAI) often focuses on users with a strong technical background, making it challenging for non-experts to understand XAI methods. This paper presents "x-[plAIn]", a new approach to make XAI more accessible to a wider audience through a custom Large Language Model (LLM), developed using ChatGPT Builder. Our goal was to design a model that can generate clear, concise summaries of various XAI methods, tailored for different audiences, including business professionals and academics. The key feature of our model is its ability to adapt explanations to match each audience group's knowledge level and interests. Our approach still offers timely insights, facilitating the decision-making process by the end users. Results from our use-case studies show that our model is effective in providing easy-to-understand, audience-specific explanations, regardless of the XAI method used. This adaptability improves the accessibility of XAI, bridging the gap between complex AI technologies and their practical applications. Our findings indicate a promising direction for LLMs in making advanced AI concepts more accessible to a diverse range of users.

We lay down the foundation of the theory of spaces of distributions on the product $X_1\times X_2$ of doubling metric measure spaces $X_1$, $X_2$ in the presence of non-negative self-adjoint operators $L_1$, $L_2$, whose heat kernels have Gaussian localization and the Markov property. This theory includes the development of two-parameter functional calculus induced by $L_1, L_2$, integral operators with highly localized kernels, test functions and distributions associated to $L_1, L_2$, spectral spaces accompanied by maximal Peetre and Nikolski type inequalities. Hardy spaces are developed in this two-parameter product setup. Two types of Besov and Triebel-Lizorkin spaces are introduced and studied: ordinary spaces and spaces with dominating mixed smoothness, with emphasis on the latter. Embedding results are obtained and spectral multiplies are developed.

In the domain of Mobility Data Science, the intricate task of interpreting models trained on trajectory data, and elucidating the spatio-temporal movement of entities, has persistently posed significant challenges. Conventional XAI techniques, although brimming with potential, frequently overlook the distinct structure and nuances inherent within trajectory data. Observing this deficiency, we introduced a comprehensive framework that harmonizes pivotal XAI techniques: LIME (Local Interpretable Model-agnostic Explanations), SHAP (SHapley Additive exPlanations), Saliency maps, attention mechanisms, direct trajectory visualization, and Permutation Feature Importance (PFI). Unlike conventional strategies that deploy these methods singularly, our unified approach capitalizes on the collective efficacy of these techniques, yielding deeper and more granular insights for models reliant on trajectory data. In crafting this synthesis, we effectively address the multifaceted essence of trajectories, achieving not only amplified interpretability but also a nuanced, contextually rich comprehension of model decisions. To validate and enhance our framework, we undertook a survey to gauge preferences and reception among various user demographics. Our findings underscored a dichotomy: professionals with academic orientations, particularly those in roles like Data Scientist, IT Expert, and ML Engineer, showcased a profound, technical understanding and often exhibited a predilection for amalgamated methods for interpretability. Conversely, end-users or individuals less acquainted with AI and Data Science showcased simpler inclinations, such as bar plots indicating timestep significance or visual depictions pinpointing pivotal segments of a vessel's trajectory.

Financial sentiment analysis plays a crucial role in decoding market trends and guiding strategic trading decisions. Despite the deployment of advanced deep learning techniques and language models to refine sentiment analysis in finance, this study breaks new ground by investigating the potential of large language models, particularly ChatGPT 3.5, in financial sentiment analysis, with a strong emphasis on the foreign exchange market (forex). Employing a zero-shot prompting approach, we examine multiple ChatGPT prompts on a meticulously curated dataset of forex-related news headlines, measuring performance using metrics such as precision, recall, f1-score, and Mean Absolute Error (MAE) of the sentiment class. Additionally, we probe the correlation between predicted sentiment and market returns as an additional evaluation approach. ChatGPT, compared to FinBERT, a well-established sentiment analysis model for financial texts, exhibited approximately 35\% enhanced performance in sentiment classification and a 36\% higher correlation with market returns. By underlining the significance of prompt engineering, particularly in zero-shot contexts, this study spotlights ChatGPT's potential to substantially boost sentiment analysis in financial applications. By sharing the utilized dataset, our intention is to stimulate further research and advancements in the field of financial services., Comment: 10 pages, 8 figures, Machine Learning with Applications (2023)

Coordination facilitation and efficient decision-making are two essential components of successful leadership. In this paper, we take an informational approach and investigate how followers' information impacts coordination and efficient leadership in a model featuring a leader and a team of followers. We show that efficiency is achieved as the unique rationalizable outcome of the game when followers possess sufficiently imprecise information. In contrast, if followers have accurate information, the leader may fail to coordinate them toward the desired outcome or even take an inefficient action herself. We discuss the implications of the results for the role of leaders in the context of financial fragility and crises.

In this paper we examine the possibility of dilute axion stars converting to photons in the weak, large-scale magnetic field of the Milky Way and show that they can resonate with the surrounding plasma and produce a sizable signal. We consider two possibilities for the plasma: free electrons and HII regions. In the former case, we argue that the frequency of the photons will be too small to be observed even by space-based radio telescopes. In the latter case, their frequency is larger, safely above the solar wind cut-off. We provide an estimate of the flux as a function of the decay constant and show that for $f_{a} < 5 \times 10^{11} \text{GeV}$, the signal will be above the radio emission of the solar system's planets and it could potentially be detected by the NCLE instrument which is on board the Chang'e-4 spacecraft. Finally, we calculate the time scale of decay of the axion star and demonstrate that back-reaction can be neglected for all physically interesting values of the decay constant, while the minimum time scale of decay is in the order of a few hours.

This paper studies the effects of unexpected changes in US monetary policy on digital asset returns. We use event study regressions and find that monetary policy surprises negatively affect BTC and ETH, the two largest digital assets, but do not significantly affect the rest of the market. Second, we use high-frequency price data to examine the effect of the FOMC statements release and Minutes release on the prices of the assets with the higher collateral usage on the Ethereum Blockchain Decentralized Finance (DeFi) ecosystem. The FOMC statement release strongly affects the volatility of digital asset returns, while the effect of the Minutes release is weaker. The volatility effect strengthened after December 2021, when the Federal Reserve changed its policy to fight inflation. We also show that some borrowing interest rates in the Ethereum DeFi ecosystem are affected positively by unexpected changes in monetary policy. In contrast, the debt outstanding and the total value locked are negatively affected. Finally, we utilize a local Ethereum Blockchain node to record the activity history of primary DeFi functions, such as depositing, borrowing, and liquidating, and study how these are influenced by the FOMC announcements over time., Comment: 33 pages, 11 figures, 9 tables

How important are leaders' actions in facilitating coordination? In this paper, we investigate their signaling role in a global games framework. A perfectly informed leader and a team of followers face a coordination problem. Despite the endogenous information generated by the leader's action, we provide a necessary and sufficient condition that makes the monotone equilibrium strategy profile uniquely $\Delta$-rationalizable and hence guarantees equilibrium uniqueness. Moreover, the unique equilibrium is fully efficient. This result remains valid when the leader observes a noisy signal about the true state except full efficiency may not be obtained. We discuss the implications of our results for a broad class of phenomena such as adoption of green technology, currency attacks and revolutions., Comment: Revised Version Uploaded

We investigate the form of the radiation emitted by an axion star-neutron star binary using a $f(r)=\text{sech(r/R)}$ profile for the axion star. Our analysis takes into account the co-rotating plasma of the neutron star. We find that that there is significant enhancement to the radiated power if the neutron star's spin is tilted towards the plane of the axion star-neutron star orbit, compared to the case where it is perpendicular. We also examine whether the neutron star's co-rotating plasma can play a role in the emitted power and we find that even though dilute axion stars can in principle radiate more efficiently than dense axion stars, they will be pulled apart by the tidal forces of the neutron star, Comment: 15 pages, 4 figures

Human-centricity is the core value behind the evolution of manufacturing towards Industry 5.0. Nevertheless, there is a lack of architecture that considers safety, trustworthiness, and human-centricity at its core. Therefore, we propose an architecture that integrates Artificial Intelligence (Active Learning, Forecasting, Explainable Artificial Intelligence), simulated reality, decision-making, and users' feedback, focusing on synergies between humans and machines. Furthermore, we align the proposed architecture with the Big Data Value Association Reference Architecture Model. Finally, we validate it on three use cases from real-world case studies.

We present PolicyCLOUD, a prototype for an extensible, serverless cloud-based system that supports evidence-based elaboration and analysis of policies. PolicyCLOUD allows flexible exploitation and management of policy-relevant dataflows by enabling the practitioner to register datasets and specify a sequence of transformations and/or information extraction through registered ingest functions. Once a possibly transformed dataset has been ingested, additional insights can be retrieved by further applying registered analytic functions. PolicyCLOUD was built as an extensible framework toward the creation of an analytic ecosystem. As of now, we developed several essential ingest and analytic functions that are built-in within the framework. They include data cleaning, enhanced interoperability, and sentiment analysis generic functions. PolicyCLOUD has also the ability to tap on the analytic capabilities of external tools. We demonstrate this with a Social Analytics tool implemented in conjunction with PolicyCLOUD and show how to benefit from policy modeling, design and simulation capabilities. Furthermore, PolicyCLOUD has developed a first of its kind legal and ethical framework that covers the usage and dissemination of datasets and analytic functions throughout its policy-relevant dataflows. The article describes and evaluates the application of PolicyCLOUD to four families of pilots that cover a wide range of policy scenarios., Comment: 18 pages + 5 reference pages

The amount and quality of datasets and tools available in the research field of hand pose and shape estimation act as evidence to the significant progress that has been made.However, even the datasets of the highest quality, reported to date, have shortcomings in annotation. We propose a refinement approach, based on differentiable ray tracing,and demonstrate how a high-quality publicly available, multi-camera dataset of hands(InterHand2.6M) can become an even better dataset, with respect to annotation quality. Differentiable ray tracing has not been employed so far to relevant problems and is hereby shown to be superior to the approximative alternatives that have been employed in the past. To tackle the lack of reliable ground truth, as far as quantitative evaluation is concerned, we resort to realistic synthetic data, to show that the improvement we induce is indeed significant. The same becomes evident in real data through visual evaluation., Comment: British Machine Vision Conference (BMVC) 2021

The implementation of Artificial Intelligence (AI) systems in the manufacturing domain enables higher production efficiency, outstanding performance, and safer operations, leveraging powerful tools such as deep learning and reinforcement learning techniques. Despite the high accuracy of these models, they are mostly considered black boxes: they are unintelligible to the human. Opaqueness affects trust in the system, a factor that is critical in the context of decision-making. We present an overview of Explainable Artificial Intelligence (XAI) techniques as a means of boosting the transparency of models. We analyze different metrics to evaluate these techniques and describe several application scenarios in the manufacturing domain., Comment: arXiv admin note: text overlap with arXiv:2102.13076 by other authors

There is a lack of a single architecture specification that addresses the needs of trusted and secure Artificial Intelligence systems with humans in the loop, such as human-centered manufacturing systems at the core of the evolution towards Industry 5.0. To realize this, we propose an architecture that integrates forecasts, Explainable Artificial Intelligence, supports collecting users' feedback, and uses Active Learning and Simulated Reality to enhance forecasts and provide decision-making recommendations. The architecture security is addressed as a general concern. We align the proposed architecture with the Big Data Value Association Reference Architecture Model. We tailor it for the domain of demand forecasting and validate it on a real-world case study.

We present HandGAN (H-GAN), a cycle-consistent adversarial learning approach implementing multi-scale perceptual discriminators. It is designed to translate synthetic images of hands to the real domain. Synthetic hands provide complete ground-truth annotations, yet they are not representative of the target distribution of real-world data. We strive to provide the perfect blend of a realistic hand appearance with synthetic annotations. Relying on image-to-image translation, we improve the appearance of synthetic hands to approximate the statistical distribution underlying a collection of real images of hands. H-GAN tackles not only the cross-domain tone mapping but also structural differences in localized areas such as shading discontinuities. Results are evaluated on a qualitative and quantitative basis improving previous works. Furthermore, we relied on the hand classification task to claim our generated hands are statistically similar to the real domain of hands., Comment: Paper accepted at The International Joint Conference on Neural Networks (IJCNN) 2021

We present a clock-driven Spiking Neural Network simulator which is up to 3x faster than the state of the art while, at the same time, being more general and requiring less programming effort on both the user's and maintainer's side. This is made possible by designing our pipeline around "work queues" which act as interfaces between stages and greatly reduce implementation complexity. We evaluate our work using three well-established SNN models on a series of benchmarks., Comment: Camera-ready version, as accepted by IJCNN 2020

Performing deep learning on end-user devices provides fast offline inference results and can help protect the user's privacy. However, running models on untrusted client devices reveals model information which may be proprietary, i.e., the operating system or other applications on end-user devices may be manipulated to copy and redistribute this information, infringing on the model provider's intellectual property. We propose the use of ARM TrustZone, a hardware-based security feature present in most phones, to confidentially run a proprietary model on an untrusted end-user device. We explore the limitations and design challenges of using TrustZone and examine potential approaches for confidential deep learning within this environment. Of particular interest is providing robust protection of proprietary model information while minimizing total performance overhead.

We develop an analytic approximation for the coincidence limit of a massive scalar propagator in an arbitrary spatially flat, homogeneous and isotropic geometry. We employ this to compute the one loop corrections to the inflaton effective potential from a quadratic coupling to a minimally coupled scalar. We also extend the Friedmann equations to cover potentials that depend locally on the Hubble parameter and the first slow roll parameter., Comment: 21 pages, 9 figures, uses LaTeX 2e. Version 2 revised for publication, including 2 new figures and 6 pages of new text to show, among other things, that our approximations apply as well to plateau potentials

We deal with homogeneous Besov and Triebel-Lizorkin spaces in the setting of a doubling metric measure space in the presence of a non-negative self-adjoint operator whose heat kernel has Gaussian localization and the Markov property. The class of almost diagonal operators on the associated sequence spaces is developed and it is shown that this class is an algebra. The boundedness of almost diagonal operators is utilized for establishing smooth molecular and atomic decompositions for the above homogeneous Besov and Triebel-Lizorkin spaces. Spectral multipliers for these spaces are established as well.

This paper presents the case study of a non-intrusive porting of a monolithic C++ library for real-time 3D hand tracking, to the domain of edge-based computation. Towards a proof of concept, the case study considers a pair of workstations, a computationally powerful and a computationally weak one. By wrapping the C++ library in Java container and by capitalizing on a Java-based offloading infrastructure that supports both CPU and GPGPU computations, we are able to establish automatically the required server-client workflow that best addresses the resource allocation problem in the effort to execute from the weak workstation. As a result, the weak workstation can perform well at the task, despite lacking the sufficient hardware to do the required computations locally. This is achieved by offloading computations which rely on GPGPU, to the powerful workstation, across the network that connects them. We show the edge-based computation challenges associated with the information flow of the ported algorithm, demonstrate how we cope with them, and identify what needs to be improved for achieving even better performance., Comment: 6 pages, 5 figures