Search

Recent high-redshift ($z>4$) spatially resolved observations with the James Webb Space Telesescope have shown the evolution of the star formation rate (SFR) surface density ($\Sigma_{\rm SFR}$) and its main sequence in the $\Sigma_{\rm SFR}$-$M_*$ diagram ($\Sigma_{\rm SFR}{\rm MS}$). The $\Sigma_{\rm SFR}{\rm MS}$\ is already observed at cosmic morning ($z\sim7.5$). The use of $\Sigma_{\rm SFR}$\ is physically motivated because it is normalized by the area in which the star formation occurs, and this indirectly considers the gas density. The $\Sigma_{\rm SFR}$-$M_*$ diagram has been shown to complement the widely used (specific) SFR-$M_*$, particularly when selecting passive galaxies. We establish the $\Sigma_{\rm SFR}$\ evolution since $z=12$ in the framework of the L-Galaxies2020 semi-analytical model (SAM), and we interpret recent observations. We estimated $\Sigma_{\rm SFR}$(-$M_*$) and the cosmic star formation rate density (CSFRD) for the simulated galaxy population and for the subsamples, which were divided into stellar mass bins in the given redshift. The simulated $\Sigma_{\rm SFR}$\ decreases by $\sim3.5$ dex from $z=12$ to $z=0$. We show that galaxies with different stellar masses have different paths of $\Sigma_{\rm SFR}$\ evolution. We find that $\Sigma_{\rm SFR}{\rm MS}$\ is already observed at $z\sim11$. The simulated $\Sigma_{\rm SFR}{\rm MS}$\ agrees with the observed one at $z=0, 1, 2, 5$, and $7.5$ and with individual galaxies at $z>10$. We show that the highest $\Sigma_{\rm SFR}{\rm MS}$\ slope of $0.709\pm0.005$ is at $z\sim3$ and decreases to $\sim0.085\pm0.003$ at $z=0$. This is mostly driven by a rapid decrease in SFR with an additional size increase for the most massive galaxies in this redshift range. This coincides with the dominance of the most massive galaxies in the CSFRD from the SAM., Comment: 7 pages, 4 figures, 2 tables. Accepted in A&A Letters

Quenching of star-formation plays a fundamental role in galaxy evolution. This process occurs due to the removal of the cold interstellar medium (ISM) or stabilization against collapse, so that gas cannot be used in the formation of new stars. In this paper, we study the effect of different mechanisms of ISM removal. In particular, we revised the well-known Baldwin-Philips-Terlevich (BPT) and $\mathrm{EW_{H\alpha}}$ vs. $\mathrm{[NII]/H\alpha}$ (WHAN) emission-line ratio diagnostics, so that we could classify all galaxies, even those not detected at some emission lines, introducing several new spectral classes. We use spectroscopic data and several physical parameters of 2409 dusty early-type galaxies in order to find out the dominant ionization source [active galactic nuclei (AGNs), young massive stars, hot low-mass evolved stars (HOLMES)] and its effect on the ISM. We find that strong AGNs can play a significant role in the ISM removal process only for galaxies with ages lower than $10^{9.4}$ yr, but we cannot rule out the influence of weak AGNs at any age. For older galaxies, HOLMES/planetary nebulae contribute significantly to the ISM removal process. Additionally, we provide the BPT and WHAN classifications not only for the selected sample but also for all 300000 galaxies in the GAMA fields., Comment: 31 pages, 16 figures, 4 tables, accepted for publication in ApJS

This paper presents an overview of the Volvo Discovery Challenge, held during the ECML-PKDD 2024 conference. The challenge's goal was to predict the failure risk of an anonymized component in Volvo trucks using a newly published dataset. The test data included observations from two generations (gen1 and gen2) of the component, while the training data was provided only for gen1. The challenge attracted 52 data scientists from around the world who submitted a total of 791 entries. We provide a brief description of the problem definition, challenge setup, and statistics about the submissions. In the section on winning methodologies, the first, second, and third-place winners of the competition briefly describe their proposed methods and provide GitHub links to their implemented code. The shared code can be interesting as an advanced methodology for researchers in the predictive maintenance domain. The competition was hosted on the Codabench platform., Comment: ECML/PKDD 2024, Discovery Challenge

Removing cold interstellar medium (ISM) from a galaxy is central to quenching star formation. However, the exact mechanism of this process remains unclear. The objective of this work is to find the mechanism responsible for dust and gas removal in simulated early-type galaxies (ETGs). A statistically significant sample of massive (M_*>$10^{10}$M$_\odot$), simulated ETG in a redshift range of 0.02--0.32 is studied in the context of its ISM properties. In particular, we investigate the cold dust and gas removal timescales, the cold gas inflows, and their relation with black hole (BH) mass. We also investigate the evolution of galaxies in the dust vs. star formation rate (SFR) plane and the influence of merger events. We find agreement with previous observational works considering the timescales of dust and HI removal from ETGs. When considering the dust-to-stellar mass ratio as a function of time in simulations, we recovered a similar decline as in the observational sample as a function of stellar age, validating its use for timing the ISM decline. Moreover, we recover the observed relation between dust mass and SFR for actively star-forming galaxies as well as for passive ETGs. We also show that starburst galaxies form their own sequence on the dust vs. SFR plot in a form $\log(M_{\rm dust, SB})= 0.913\times \log({\rm SFR}) + 6.533$ with $2\sigma$ scatter of 0.32. Finally, we find that type II supernova reverse shocks dominate the dust destruction at the early stages of ETG evolution, while at later times stellar feedback becomes more important. We show that merger events lead to morphological transformations by increasing the bulge-to-total stellar mass ratio followed by an increase in BH masses. The BH feedback resulting from radio mode accretion prevents the hot halo gas from cooling, indirectly leading to a decrease in the SFR., Comment: 15 pages, 12 figures, accepted in A&A

Unsupervised meta-learning aims to learn feature representations from unsupervised datasets that can transfer to downstream tasks with limited labeled data. In this paper, we propose a novel approach to unsupervised meta-learning that leverages the generalization abilities of in-context learning observed in transformer architectures. Our method reframes meta-learning as a sequence modeling problem, enabling the transformer encoder to learn task context from support images and utilize it to predict query images. At the core of our approach lies the creation of diverse tasks generated using a combination of data augmentations and a mixing strategy that challenges the model during training while fostering generalization to unseen tasks at test time. Experimental results on benchmark datasets showcase the superiority of our approach over existing unsupervised meta-learning baselines, establishing it as the new state-of-the-art in the field. Remarkably, our method achieves competitive results with supervised and self-supervised approaches, underscoring the efficacy of the model in leveraging generalization over memorization.

Exploring the missing values is an essential but challenging issue due to the complex latent spatio-temporal correlation and dynamic nature of time series. Owing to the outstanding performance in dealing with structure learning potentials, Graph Neural Networks (GNNs) and Recurrent Neural Networks (RNNs) are often used to capture such complex spatio-temporal features in multivariate time series. However, these data-driven models often fail to capture the essential spatio-temporal relationships when significant signal corruption occurs. Additionally, calculating the high-order neighbor nodes in these models is of high computational complexity. To address these problems, we propose a novel higher-order spatio-temporal physics-incorporated GNN (HSPGNN). Firstly, the dynamic Laplacian matrix can be obtained by the spatial attention mechanism. Then, the generic inhomogeneous partial differential equation (PDE) of physical dynamic systems is used to construct the dynamic higher-order spatio-temporal GNN to obtain the missing time series values. Moreover, we estimate the missing impact by Normalizing Flows (NF) to evaluate the importance of each node in the graph for better explainability. Experimental results on four benchmark datasets demonstrate the effectiveness of HSPGNN and the superior performance when combining various order neighbor nodes. Also, graph-like optical flow, dynamic graphs, and missing impact can be obtained naturally by HSPGNN, which provides better dynamic analysis and explanation than traditional data-driven models. Our code is available at https://github.com/gorgen2020/HSPGNN., Comment: 18 pages, 7 figures, CIKM 2024

Evolutionary Algorithms (EAs) are often challenging to apply in real-world settings since evolutionary computations involve a large number of evaluations of a typically expensive fitness function. For example, an evaluation could involve training a new machine learning model. An approximation (also known as meta-model or a surrogate) of the true function can be used in such applications to alleviate the computation cost. In this paper, we propose a two-stage surrogate-assisted evolutionary approach to address the computational issues arising from using Genetic Algorithm (GA) for feature selection in a wrapper setting for large datasets. We define 'Approximation Usefulness' to capture the necessary conditions to ensure correctness of the EA computations when an approximation is used. Based on this definition, we propose a procedure to construct a lightweight qualitative meta-model by the active selection of data instances. We then use a meta-model to carry out the feature selection task. We apply this procedure to the GA-based algorithm CHC (Cross generational elitist selection, Heterogeneous recombination and Cataclysmic mutation) to create a Qualitative approXimations variant, CHCQX. We show that CHCQX converges faster to feature subset solutions of significantly higher accuracy (as compared to CHC), particularly for large datasets with over 100K instances. We also demonstrate the applicability of the thinking behind our approach more broadly to Swarm Intelligence (SI), another branch of the Evolutionary Computation (EC) paradigm with results of PSOQX, a qualitative approximation adaptation of the Particle Swarm Optimization (PSO) method. A GitHub repository with the complete implementation is available.

This paper investigates how various randomization techniques impact Deep Neural Networks (DNNs). Randomization, like weight noise and dropout, aids in reducing overfitting and enhancing generalization, but their interactions are poorly understood. The study categorizes randomness techniques into four types and proposes new methods: adding noise to the loss function and random masking of gradient updates. Using Particle Swarm Optimizer (PSO) for hyperparameter optimization, it explores optimal configurations across MNIST, FASHION-MNIST, CIFAR10, and CIFAR100 datasets. Over 30,000 configurations are evaluated, revealing data augmentation and weight initialization randomness as main performance contributors. Correlation analysis shows different optimizers prefer distinct randomization types. The complete implementation and dataset are available on GitHub.

To meet the urgent requirements for the climate change mitigation, several proactive measures of energy efficiency have been implemented in maritime industry. Many of these practices depend highly on the onboard data of vessel's operation and environmental conditions. In this paper, a high resolution onboard data from passenger vessels in short-sea shipping (SSS) have been collected and preprocessed. We first investigated the available data to deploy it effectively to model the physics of the vessel, and hence the vessel performance. Since in SSS, the weather measurements and forecasts might have not been in temporal and spatial resolutions that accurately representing the actual environmental conditions. Then, We proposed a data-driven modeling approach for vessel energy efficiency. This approach addresses the challenges of data representation and energy modeling by combining and aggregating data from multiple sources and seamlessly integrates explainable artificial intelligence (XAI) to attain clear insights about the energy efficiency for a vessel in SSS. After that, the developed model of energy efficiency has been utilized in developing a framework for optimizing the vessel voyage to minimize the fuel consumption and meeting the constraint of arrival time. Moreover, we developed a spatial clustering approach for labeling the vessel paths to detect the paths for vessels with operating routes of repeatable and semi-repeatable paths., Comment: 20 pages, 28 figures, 9 tables, Keywords: Short-sea shipping, energy efficiency, spatiotemporal aggregation, voyage speed optimization, Time-series analysis, spatial clustering, vessel path identification, maritime transportation. arXiv admin note: substantial text overlap with arXiv:2402.00698, arXiv:2403.05778

This paper addresses the challenge of identifying the paths for vessels with operating routes of repetitive paths, partially repetitive paths, and new paths. We propose a spatial clustering approach for labeling the vessel paths by using only position information. We develop a path clustering framework employing two methods: a distance-based path modeling and a likelihood estimation method. The former enhances the accuracy of path clustering through the integration of unsupervised machine learning techniques, while the latter focuses on likelihood-based path modeling and introduces segmentation for a more detailed analysis. The result findings highlight the superior performance and efficiency of the developed approach, as both methods for clustering vessel paths into five classes achieve a perfect F1-score. The approach aims to offer valuable insights for route planning, ultimately contributing to improving safety and efficiency in maritime transportation., Comment: Keywords: Spatial clustering, vessel path identification, maritime transportation, average nearest neighbor distance, hierarchical clustering, likelihood estimation. This preprint has 12 pages, 14 figures, 2 tables

Several approaches have been developed for improving the ship energy efficiency, thereby reducing operating costs and ensuring compliance with climate change mitigation regulations. Many of these approaches will heavily depend on measured data from onboard IoT devices, including operational and environmental information, as well as external data sources for additional navigational data. In this paper, we develop a framework that implements time-series analysis techniques to optimize the vessel's speed profile for improving the vessel's energy efficiency. We present a case study involving a real-world data from a passenger vessel that was collected over a span of 15 months in the south of Sweden. The results indicate that the implemented models exhibit a range of outcomes and adaptability across different scenarios. The findings highlight the effectiveness of time-series analysis approach for optimizing vessel voyages within the context of constrained landscapes, as often seen in short-sea shipping., Comment: 8 pages, 7 figures and four tables

This paper investigates the issue of privacy in a learning scenario where users share knowledge for a recommendation task. Our study contributes to the growing body of research on privacy-preserving machine learning and underscores the need for tailored privacy techniques that address specific attack patterns rather than relying on one-size-fits-all solutions. We use the latent bandit setting to evaluate the trade-off between privacy and recommender performance by employing various aggregation strategies, such as averaging, nearest neighbor, and clustering combined with noise injection. More specifically, we simulate a linkage attack scenario leveraging publicly available auxiliary information acquired by the adversary. Our results on three open real-world datasets reveal that adding noise using the Laplace mechanism to an individual user's data record is a poor choice. It provides the highest regret for any noise level, relative to de-anonymization probability and the ADS metric. Instead, one should combine noise with appropriate aggregation strategies. For example, using averages from clusters of different sizes provides flexibility not achievable by varying the amount of noise alone. Generally, no single aggregation strategy can consistently achieve the optimum regret for a given desired level of privacy.

Federated Learning (FL) allows several clients to construct a common global machine-learning model without having to share their data. FL, however, faces the challenge of statistical heterogeneity between the client's data, which degrades performance and slows down the convergence toward the global model. In this paper, we provide theoretical proof that minimizing heterogeneity between clients facilitates the convergence of a global model for every single client. This becomes particularly important under empirical concept shifts among clients, rather than merely considering imbalanced classes, which have been studied until now. Therefore, we propose a method for knowledge transfer between clients where the server trains client-specific generators. Each generator generates samples for the corresponding client to remove the conflict with other clients' models. Experiments conducted on synthetic and real data, along with a theoretical study, support the effectiveness of our method in constructing a well-generalizable global model by reducing the conflict between local models.

Transient absorption (TA) spectroscopy has proved fundamental to our understanding of energy and charge transfer in biological systems, allowing measurements of photoactive proteins on sub-picosecond timescales. Recently, ultrafast TA spectroscopy has been applied in vivo, providing sub-picosecond measurements of photosynthetic light harvesting and electron transfer processes within living photosynthetic microorganisms. The analysis of the resultant data is hindered by the number of different photoactive pigments and the associated complexity of photoactive reaction schemes within living cells. Here we show how in vivo ultrafast TA spectroscopy can be applied to a diverse array of organisms from the tree of life, both photosynthetic and non-photosynthetic. We have developed a series of software tools for performing global, lifetime and target analysis of in vivo TA datasets. These advances establish in vivo TA spectroscopy as a versatile technique for studying energy and charge transfer in living systems., Comment: 39 pages

Explainable Artificial Intelligence (XAI) fills the role of a critical interface fostering interactions between sophisticated intelligent systems and diverse individuals, including data scientists, domain experts, end-users, and more. It aids in deciphering the intricate internal mechanisms of ``black box'' Machine Learning (ML), rendering the reasons behind their decisions more understandable. However, current research in XAI primarily focuses on two aspects; ways to facilitate user trust, or to debug and refine the ML model. The majority of it falls short of recognising the diverse types of explanations needed in broader contexts, as different users and varied application areas necessitate solutions tailored to their specific needs. One such domain is Predictive Maintenance (PdM), an exploding area of research under the Industry 4.0 \& 5.0 umbrella. This position paper highlights the gap between existing XAI methodologies and the specific requirements for explanations within industrial applications, particularly the Predictive Maintenance field. Despite explainability's crucial role, this subject remains a relatively under-explored area, making this paper a pioneering attempt to bring relevant challenges to the research community's attention. We provide an overview of predictive maintenance tasks and accentuate the need and varying purposes for corresponding explanations. We then list and describe XAI techniques commonly employed in the literature, discussing their suitability for PdM tasks. Finally, to make the ideas and claims more concrete, we demonstrate XAI applied in four specific industrial use cases: commercial vehicles, metro trains, steel plants, and wind farms, spotlighting areas requiring further research., Comment: 51 pages, 9 figures

A fundamental feature of human intelligence is the ability to infer high-level abstractions from low-level sensory data. An essential component of such inference is the ability to discover modularized generative mechanisms. Despite many efforts to use statistical learning and pattern recognition for finding disentangled factors, arguably human intelligence remains unmatched in this area. In this paper, we investigate a problem of learning, in a fully unsupervised manner, the inverse of a set of independent mechanisms from distorted data points. We postulate, and justify this claim with experimental results, that an important weakness of existing machine learning solutions lies in the insufficiency of cross-module diversification. Addressing this crucial discrepancy between human and machine intelligence is an important challenge for pattern recognition systems. To this end, our work proposes an unsupervised method that discovers and disentangles a set of independent mechanisms from unlabeled data, and learns how to invert them. A number of experts compete against each other for individual data points in an adversarial setting: one that best inverses the (unknown) generative mechanism is the winner. We demonstrate that introducing an orthogonalization layer into the expert architectures enforces additional diversity in the outputs, leading to significantly better separability. Moreover, we propose a procedure for relocating data points between experts to further prevent any one from claiming multiple mechanisms. We experimentally illustrate that these techniques allow discovery and modularization of much less pronounced transformations, in addition to considerably faster convergence.

Graph neural networks (GNNs), especially dynamic GNNs, have become a research hotspot in spatio-temporal forecasting problems. While many dynamic graph construction methods have been developed, relatively few of them explore the causal relationship between neighbour nodes. Thus, the resulting models lack strong explainability for the causal relationship between the neighbour nodes of the dynamically generated graphs, which can easily lead to a risk in subsequent decisions. Moreover, few of them consider the uncertainty and noise of dynamic graphs based on the time series datasets, which are ubiquitous in real-world graph structure networks. In this paper, we propose a novel Dynamic Diffusion-Variational Graph Neural Network (DVGNN) for spatio-temporal forecasting. For dynamic graph construction, an unsupervised generative model is devised. Two layers of graph convolutional network (GCN) are applied to calculate the posterior distribution of the latent node embeddings in the encoder stage. Then, a diffusion model is used to infer the dynamic link probability and reconstruct causal graphs in the decoder stage adaptively. The new loss function is derived theoretically, and the reparameterization trick is adopted in estimating the probability distribution of the dynamic graphs by Evidence Lower Bound during the backpropagation period. After obtaining the generated graphs, dynamic GCN and temporal attention are applied to predict future states. Experiments are conducted on four real-world datasets of different graph structures in different domains. The results demonstrate that the proposed DVGNN model outperforms state-of-the-art approaches and achieves outstanding Root Mean Squared Error result while exhibiting higher robustness. Also, by F1-score and probability distribution analysis, we demonstrate that DVGNN better reflects the causal relationship and uncertainty of dynamic graphs.

In the latent bandit problem, the learner has access to reward distributions and -- for the non-stationary variant -- transition models of the environment. The reward distributions are conditioned on the arm and unknown latent states. The goal is to use the reward history to identify the latent state, allowing for the optimal choice of arms in the future. The latent bandit setting lends itself to many practical applications, such as recommender and decision support systems, where rich data allows the offline estimation of environment models with online learning remaining a critical component. Previous solutions in this setting always choose the highest reward arm according to the agent's beliefs about the state, not explicitly considering the value of information-gathering arms. Such information-gathering arms do not necessarily provide the highest reward, thus may never be chosen by an agent that chooses the highest reward arms at all times. In this paper, we present a method for information-gathering in latent bandits. Given particular reward structures and transition matrices, we show that choosing the best arm given the agent's beliefs about the states incurs higher regret. Furthermore, we show that by choosing arms carefully, we obtain an improved estimation of the state distribution, and thus lower the cumulative regret through better arm choices in the future. We evaluate our method on both synthetic and real-world data sets, showing significant improvement in regret over state-of-the-art methods.

Photosystems II and I (PSII and PSI) are the reaction centre complexes that drive the light reactions of photosynthesis. PSII performs light-driven water oxidation (quantum efficiencies and catalysis rates of up to 80% and 1000 $e^{-}\text{s}^{-1}$, respectively) and PSI further photo-energises the harvested electrons (quantum efficiencies of ~100%). The impressive performance of the light harvesting components of photosynthesis has motivated extensive biological, artificial and biohybrid approaches to re-wire photosynthesis to enable higher efficiencies and new reaction pathways, such as H2 evolution or alternative CO2 fixation. To date these approaches have focussed on charge extraction at the terminal electron quinones of PSII and terminal iron-sulfur clusters of PSI. Ideally electron extraction would be possible immediately from the photoexcited reaction centres to enable the greatest thermodynamic gains. However, this was believed to be impossible because the reaction centres are buried around 4 nm within PSII and 5 nm within PSI from the cytoplasmic face. Here, we demonstrate using in vivo ultrafast transient absorption (TA) spectroscopy that it is possible to extract electrons directly from photoexcited PSI and PSII, using both live cyanobacterial cells and isolated photosystems, with the exogenous electron mediator 2,6-dichloro1,4-benzoquinone (DCBQ). We postulate that DCBQ can oxidise peripheral chlorophyll pigments participating in highly delocalised charge transfer (CT) states after initial photoexcitation. Our results open new avenues to study and re-wire photosynthesis for bioenergy and semi-artificial photosynthesis.