Search

Evolutionary dynamics are shaped by a variety of fundamental, generic drivers, including spatial structure, ecology, and selection pressure. These drivers impact the trajectory of evolution, and have been hypothesized to influence phylogenetic structure. Here, we set out to assess (1) if spatial structure, ecology, and selection pressure leave detectable signatures in phylogenetic structure, (2) the extent, in particular, to which ecology can be detected and discerned in the presence of spatial structure, and (3) the extent to which these phylogenetic signatures generalize across evolutionary systems. To this end, we analyze phylogenies generated by manipulating spatial structure, ecology, and selection pressure within three computational models of varied scope and sophistication. We find that selection pressure, spatial structure, and ecology have characteristic effects on phylogenetic metrics, although these effects are complex and not always intuitive. Signatures have some consistency across systems when using equivalent taxonomic unit definitions (e.g., individual, genotype, species). Further, we find that sufficiently strong ecology can be detected in the presence of spatial structure. We also find that, while low-resolution phylogenetic reconstructions can bias some phylogenetic metrics, high-resolution reconstructions recapitulate them faithfully. Although our results suggest potential for evolutionary inference of spatial structure, ecology, and selection pressure through phylogenetic analysis, further methods development is needed to distinguish these drivers' phylometric signatures from each other and to appropriately normalize phylogenetic metrics. With such work, phylogenetic analysis could provide a versatile toolkit to study large-scale evolving populations.

Continuing generation of novelty, complexity, and adaptation are well-established as core aspects of open-ended evolution. However, it has yet to be firmly established to what extent these phenomena are coupled and by what means they interact. In this work, we track the co-evolution of novelty, complexity, and adaptation in a case study from the DISHTINY simulation system, which is designed to study the evolution of digital multicellularity. In this case study, we describe ten qualitatively distinct multicellular morphologies, several of which exhibit asymmetrical growth and distinct life stages. We contextualize the evolutionary history of these morphologies with measurements of complexity and adaptation. Our case study suggests a loose -- sometimes divergent -- relationship can exist among novelty, complexity, and adaptation.

The paper aims to investigate relevant computational issues of deep neural network architectures with an eye to the interaction between the optimization algorithm and the classification performance. In particular, we aim to analyze the behaviour of state-of-the-art optimization algorithms in relationship to their hyperparameters setting in order to detect robustness with respect to the choice of a certain starting point in ending on different local solutions. We conduct extensive computational experiments using nine open-source optimization algorithms to train deep Convolutional Neural Network architectures on an image multi-class classification task. Precisely, we consider several architectures by changing the number of layers and neurons per layer, in order to evaluate the impact of different width and depth structures on the computational optimization performance.

In the realm of financial analytics, leveraging unstructured data, such as earnings conference calls (ECCs), to forecast stock volatility is a critical challenge that has attracted both academics and investors. While previous studies have used multimodal deep learning-based models to obtain a general view of ECCs for volatility predicting, they often fail to capture detailed, complex information. Our research introduces a novel framework: \textbf{ECC Analyzer}, which utilizes large language models (LLMs) to extract richer, more predictive content from ECCs to aid the model's prediction performance. We use the pre-trained large models to extract textual and audio features from ECCs and implement a hierarchical information extraction strategy to extract more fine-grained information. This strategy first extracts paragraph-level general information by summarizing the text and then extracts fine-grained focus sentences using Retrieval-Augmented Generation (RAG). These features are then fused through multimodal feature fusion to perform volatility prediction. Experimental results demonstrate that our model outperforms traditional analytical benchmarks, confirming the effectiveness of advanced LLM techniques in financial analysis., Comment: 9 pages, 1 figures, 2 tables

The Lunar Gravitational-wave Antenna (LGWA) is a proposed array of next-generation inertial sensors to monitor the response of the Moon to gravitational waves (GWs). Given the size of the Moon and the expected noise produced by the lunar seismic background, the LGWA would be able to observe GWs from about 1 mHz to 1 Hz. This would make the LGWA the missing link between space-borne detectors like LISA with peak sensitivities around a few millihertz and proposed future terrestrial detectors like Einstein Telescope or Cosmic Explorer. In this article, we provide a first comprehensive analysis of the LGWA science case including its multi-messenger aspects and lunar science with LGWA data. We also describe the scientific analyses of the Moon required to plan the LGWA mission.

The integration of Artificial Intelligence (AI) techniques, particularly large language models (LLMs), in finance has garnered increasing academic attention. Despite progress, existing studies predominantly focus on tasks like financial text summarization, question-answering (Q$\&$A), and stock movement prediction (binary classification), with a notable gap in the application of LLMs for financial risk prediction. Addressing this gap, in this paper, we introduce \textbf{RiskLabs}, a novel framework that leverages LLMs to analyze and predict financial risks. RiskLabs uniquely combines different types of financial data, including textual and vocal information from Earnings Conference Calls (ECCs), market-related time series data, and contextual news data surrounding ECC release dates. Our approach involves a multi-stage process: initially extracting and analyzing ECC data using LLMs, followed by gathering and processing time-series data before the ECC dates to model and understand risk over different timeframes. Using multimodal fusion techniques, RiskLabs amalgamates these varied data features for comprehensive multi-task financial risk prediction. Empirical experiment results demonstrate RiskLab's effectiveness in forecasting both volatility and variance in financial markets. Through comparative experiments, we demonstrate how different data sources contribute to financial risk assessment and discuss the critical role of LLMs in this context. Our findings not only contribute to the AI in finance application but also open new avenues for applying LLMs in financial risk assessment., Comment: 24 pages, 7 figures, 5 tables, 1 algorithm

This paper presents the proof-of-concept of the first automatic voice assistant ever built in Wolof language, the main vehicular language spoken in Senegal. This voicebot is the result of a collaborative research project between Orange Innovation in France, Orange Senegal (aka Sonatel) and ADNCorp, a small IT company based in Dakar, Senegal. The purpose of the voicebot is to provide information to Orange customers about the Sargal loyalty program of Orange Senegal by using the most natural mean to communicate: speech. The voicebot receives in input the customer's oral request that is then processed by a SLU system to reply to the customer's request using audio recordings. The first results of this proof-of-concept are encouraging as we achieved 22\% of WER for the ASR task and 78\% of F1-score on the NLU task., Comment: in French language

Depth estimation is a fundamental knowledge for autonomous systems that need to assess their own state and perceive the surrounding environment. Deep learning algorithms for depth estimation have gained significant interest in recent years, owing to the potential benefits of this methodology in overcoming the limitations of active depth sensing systems. Moreover, due to the low cost and size of monocular cameras, researchers have focused their attention on monocular depth estimation (MDE), which consists in estimating a dense depth map from a single RGB video frame. State of the art MDE models typically rely on vision transformers (ViT) architectures that are highly deep and complex, making them unsuitable for fast inference on devices with hardware constraints. Purposely, in this paper, we address the problem of exploiting ViT in MDE on embedded devices. Those systems are usually characterized by limited memory capabilities and low-power CPU/GPU. We propose METER, a novel lightweight vision transformer architecture capable of achieving state of the art estimations and low latency inference performances on the considered embedded hardwares: NVIDIA Jetson TX1 and NVIDIA Jetson Nano. We provide a solution consisting of three alternative configurations of METER, a novel loss function to balance pixel estimation and reconstruction of image details, and a new data augmentation strategy to improve the overall final predictions. The proposed method outperforms previous lightweight works over the two benchmark datasets: the indoor NYU Depth v2 and the outdoor KITTI.

Ground-truth RGBD data are fundamental for a wide range of computer vision applications; however, those labeled samples are difficult to collect and time-consuming to produce. A common solution to overcome this lack of data is to employ graphic engines to produce synthetic proxies; however, those data do not often reflect real-world images, resulting in poor performance of the trained models at the inference step. In this paper we propose a novel training pipeline that incorporates Diffusion4D (D4D), a customized 4-channels diffusion model able to generate realistic RGBD samples. We show the effectiveness of the developed solution in improving the performances of deep learning models on the monocular depth estimation task, where the correspondence between RGB and depth map is crucial to achieving accurate measurements. Our supervised training pipeline, enriched by the generated samples, outperforms synthetic and original data performances achieving an RMSE reduction of (8.2%, 11.9%) and (8.1%, 6.1%) respectively on the indoor NYU Depth v2 and the outdoor KITTI dataset.

Data stream algorithms tackle operations on high-volume sequences of read-once data items. Data stream scenarios include inherently real-time systems like sensor networks and financial markets. They also arise in purely-computational scenarios like ordered traversal of big data or long-running iterative simulations. In this work, we develop methods to maintain running archives of stream data that are temporally representative, a task we call "stream curation." Our approach contributes to rich existing literature on data stream binning, which we extend by providing stateless (i.e., non-iterative) curation schemes that enable key optimizations to trim archive storage overhead and streamline processing of incoming observations. We also broaden support to cover new trade-offs between curated archive size and temporal coverage. We present a suite of five stream curation algorithms that span $\mathcal{O}(n)$, $\mathcal{O}(\log n)$, and $\mathcal{O}(1)$ orders of growth for retained data items. Within each order of growth, algorithms are provided to maintain even coverage across history or bias coverage toward more recent time points. More broadly, memory-efficient stream curation can boost the data stream mining capabilities of low-grade hardware in roles such as sensor nodes and data logging devices.

Since the advent of modern bioinformatics, the challenging, multifaceted problem of reconstructing phylogenetic history from biological sequences has hatched perennial statistical and algorithmic innovation. Studies of the phylogenetic dynamics of digital, agent-based evolutionary models motivate a peculiar converse question: how to best engineer tracking to facilitate fast, accurate, and memory-efficient lineage reconstructions? Here, we formally describe procedures for phylogenetic analysis in both serial and distributed computing scenarios. With respect to the former, we demonstrate reference-counting-based pruning of extinct lineages. For the latter, we introduce a trie-based phylogenetic reconstruction approach for "hereditary stratigraphy" genome annotations. This process allows phylogenetic relationships between genomes to be inferred by comparing their similarities, akin to reconstruction of natural history from biological DNA sequences. Phylogenetic analysis capabilities significantly advance distributed agent-based simulations as a tool for evolutionary research, and also benefit application-oriented evolutionary computing. Such tracing could extend also to other digital artifacts that proliferate through replication, like digital media and computer viruses.

This paper presents BioNeRF, a biologically plausible architecture that models scenes in a 3D representation and synthesizes new views through radiance fields. Since NeRF relies on the network weights to store the scene's 3-dimensional representation, BioNeRF implements a cognitive-inspired mechanism that fuses inputs from multiple sources into a memory-like structure, improving the storing capacity and extracting more intrinsic and correlated information. BioNeRF also mimics a behavior observed in pyramidal cells concerning contextual information, in which the memory is provided as the context and combined with the inputs of two subsequent neural models, one responsible for producing the volumetric densities and the other the colors used to render the scene. Experimental results show that BioNeRF outperforms state-of-the-art results concerning a quality measure that encodes human perception in two datasets: real-world images and synthetic data.

We consider the one-loop effective vertex for the interaction of a gluon with a Reggeized gluon and a Higgs boson in the infinite-top-mass limit, which is described by a dimension-5 non-renormalizable operator. This vertex enters the calculation of differential cross sections for the forward inclusive production of a Higgs boson in high-energy proton-proton collisions, possibly in association with a backward jet or identified hadron, in a framework where next-to-leading logarithms of the energy are resummed to all orders. The effective vertex is extracted from the high-energy behavior of two-to-two amplitudes for the Higgs production in parton-parton collisions and relies on the validity of the Regge form for these amplitudes. We find that the usual eikonal approximation (Gribov prescription) for the Regge limit and the known region-expansion technique in this limit lead to an incomplete result for the amplitude. The discrepancy is traced back to the non-renormalizable nature of the involved operator. However, the Regge limit of the exact QCD amplitude agrees with the Regge-pole exchange form at one loop, nontrivially supporting the Reggeization hypothesis., Comment: 46 pages, 12 Figures, version published in JHEP

We present a new method and implementation to obtain Bayesian posteriors on the amplitude parameters $\{h_0, \cos \iota, \psi, \phi_0\}$ of continuous-gravitational waves emitted by known pulsars. This approach leverages the well-established $\mathcal{F}$-statistic framework and software. We further explore the benefits of employing a likelihood function that is analytically marginalized over $\phi_0$, which avoids signal degeneracy problems in the $\psi$-$\phi_0$ subspace. The method is tested on simulated signals, hardware injections in Advanced-LIGO detector data, and by performing percentile-percentile (PP) self-consistency tests of the posteriors via Monte-Carlo simulations. We apply our methodology to PSR J1526-2744, a recently discovered millisecond pulsar. We find no evidence for a signal and obtain a Bayesian upper limit $h_0^{95\%}$ on the gravitational-wave amplitude of approximately $7 \times 10^{-27}$, comparable with a previous frequentist upper limit., Comment: 15 pages, 17 figures, updated to match published version

Local zoning ordinances across the United States have the impact of restricting development of energy infrastructure, including utility-scale solar photovoltaics. While these ordinances may be developed for legitimate purposes to protect public health and safety, they could impede or increase costs of power sector decarbonization. We quantify the role of utility-scale solar zoning ordinances on power sector decarbonization across the Great Lakes region (Illinois, Indiana, Michigan, Minnesota, Ohio, and Wisconsin) by integrating 6,300 rural community zoning ordinances into a power system planning model. Relative to no ordinances, solar zoning ordinances reduce total potential deployment of solar PV by 52% (or 1.6 TW) across our region. Currently, however, the biggest zoning barrier to deployment is zoning ordinances which are silent on utility-scale solar. Deployment restrictions translate to up to 4 GW greater investment needs and 5.6% greater PV investment costs to achieve a 10% PV generation target. Starker shifts occur at the state level, e.g. Wisconsin sees a 40% reduction in PV investments due to zoning restrictions. Our results underscore the need for planning that aligns local zoning laws with state and regional goals.

We present the early release of the atlas of continuous gravitational waves covering frequencies from 20 Hz to 1500 Hz and spindowns from -5e-10 to 5e-10 Hz/s. Compared to the previous atlas release we have greatly expanded the parameter space, and we now also provide polarization-specific data - both for signal-to-noise ratios and for the upper limits. Continuous wave searches are computationally difficult and take a long time to complete. The atlas enables new searches to be performed using modest computing power. To allow new searches to start sooner, we are releasing this data early, before our followup stages have completed., Comment: arXiv admin note: text overlap with arXiv:2202.10598

Here, the synthesis of polyglobalide (PGl) by enzymatic ring-opening polymerization (e-ROP) is investigated, using pressurized carbon dioxide (CO2), pressurized CO2 + dichloromethane (DCM), and pressurized propane as solvents. Particularly, the effects of phase equilibrium on the course of e-ROP and PGl final properties are discussed. The partition coefficients of CO2, DCM, propane, globalide and PGl were calculated with help of thermodynamic models, providing proper understanding of monomer partitioning in the reaction system. Reactions performed in pure CO2 resulted in monomer conversions of 100%. Besides, when only one liquid phase was present inside the reactor, PGl samples presented low polydispersities and high average molecular weights. When carried out in CO2 + DCM, e-ROP resulted in lower monomer conversions and PGl samples with higher polydispersities and lower average molecular weights. Finally, reactions carried out in pressurized propane (200 bar) produced PGl samples with the highest average molecular weights among the analyzed products., Comment: 41 pages, 9 figures, 8 tables

Photonic integrated circuits are finding use in a variety of applications including optical transceivers, LIDAR, bio-sensing, photonic quantum computing, and Machine Learning (ML). In particular, with the exponentially increasing sizes of ML models, photonics-based accelerators are getting special attention as a sustainable solution because they can perform ML inferences with multiple orders of magnitude higher energy efficiency than CMOS-based accelerators. However, recent studies have shown that hardware manufacturing and infrastructure contribute significantly to the carbon footprint of computing devices, even surpassing the emissions generated during their use. For example, the manufacturing process accounts for 74% of the total carbon emissions from Apple in 2019. This prompts us to ask -- if we consider both the embodied (manufacturing) and operational carbon cost of photonics, is it indeed a viable avenue for a sustainable future? So, in this paper, we build a carbon footprint model for photonic chips and investigate the sustainability of photonics-based accelerators by conducting a case study on ADEPT, a photonics-based accelerator for deep neural network inference. Our analysis shows that photonics can reduce both operational and embodied carbon footprints with its high energy efficiency and at least 4$\times$ less fabrication carbon cost per unit area than 28 nm CMOS.

Large Language Models (LLMs) are increasingly bringing advances to Natural Language Processing. However, low-resource languages, those lacking extensive prominence in datasets for various NLP tasks, or where existing datasets are not as substantial, such as Portuguese, already obtain several benefits from LLMs, but not to the same extent. LLMs trained on multilingual datasets normally struggle to respond to prompts in Portuguese satisfactorily, presenting, for example, code switching in their responses. This work proposes a fine-tuned LLaMA 2-based model for Portuguese prompts named Bode in two versions: 7B and 13B. We evaluate the performance of this model in classification tasks using the zero-shot approach with in-context learning, and compare it with other LLMs. Our main contribution is to bring an LLM with satisfactory results in the Portuguese language, as well as to provide a model that is free for research or commercial purposes., Comment: 10 pages, 3 figures

Pulmonary nodules may be an early manifestation of lung cancer, the leading cause of cancer-related deaths among both men and women. Numerous studies have established that deep learning methods can yield high-performance levels in the detection of lung nodules in chest X-rays. However, the lack of gold-standard public datasets slows down the progression of the research and prevents benchmarking of methods for this task. To address this, we organized a public research challenge, NODE21, aimed at the detection and generation of lung nodules in chest X-rays. While the detection track assesses state-of-the-art nodule detection systems, the generation track determines the utility of nodule generation algorithms to augment training data and hence improve the performance of the detection systems. This paper summarizes the results of the NODE21 challenge and performs extensive additional experiments to examine the impact of the synthetically generated nodule training images on the detection algorithm performance., Comment: 15 pages, 5 figures