Search

We derive a novel BPS bound from chiral perturbation theory minimally coupled to electrodynamics at finite isospin chemical potential. At a critical value of the isospin chemical potential, a system of three first-order differential field equations (which implies the second-order field equations) for the gauge field and the hadronic profile can be derived from the requirement to saturate the bound. These BPS configurations represent magnetic multi-vortices with quantized flux supported by a superconducting current. The corresponding topological charge density is related to the magnetic flux density, but is screened by the hadronic profile. Such a screening effect allows to derive the maximal value of the magnetic field generated by these BPS magnetic vortices, being $B_{max}=2,04 \times 10^{14}G$. The solution for a single BPS vortex is discussed in detail, and some physical consequences, together with the comparison with the magnetic vortices in the Ginzburg-Landau theory at critical coupling, are described., Comment: V3: Presentation improved, discussion on the interaction of vortices with quarks together with a more detailed comparison with the existing literature included. Version accepted for publication on JHEP

We present a new magnetic-atmosphere model code for obtaining synthetic spectral fluxes of hydrogen-rich magnetic white dwarfs. To date, observed spectra have been analyzed with models that neglect the magnetic field's effects on the atomic populations. In this work, we incorporate state-of-art theory in the evaluation of numerical densities of atoms, free electrons, and ions in local thermodynamical equilibrium under the action of a magnetic field. The energy distribution of atoms is rigorously evaluated for arbitrary field strength. This energy pattern includes going from tightly bound states to metastable or truly bound, highly excited states embedded in the continuum, that is, over the first Landau level. Finite nuclear mass effects and the coupling between the internal atomic structure and the motion of the atom across the magnetic field are also considered. Synthetic fluxes are generated with integrations of numerical solutions of polarized radiative transfer over the visible stellar disk using a spherical t-design method. The atmosphere code is tested with observations from the Sloan Digital Sky Survey for a group of known magnetic white dwarfs. Physical stellar parameters are obtained from least-squares fits to the observed energy distribution and compared with results of previous works. We show that the use of zerofield ionization equilibrium in spectral analyses can lead to underestimated effective temperatures for highly magnetic white dwarfs., Comment: 14 pages, 11 figures. Accepted for publication in Astronomy & Astrophysics

Approaching thermodynamic limits in light harvesting requires enabling nonreciprocal thermal emission. The majority of previously reported nonreciprocal thermal emitters operate in reflection mode, following original proposals by M. Green and others. In these proposals, cascaded nonreciprocal junctions that re-direct each junction's emission towards a subsequent one are employed for efficient light-harvesting. Recently, simplified concepts have been proposed in solar photovoltaics and thermophotovoltaics that leverage the concept of tandem junctions to approach thermodynamic limits. In these simplified scenarios, polarization-independent nonreciprocal response in transmission mode is required. We propose a pattern-free heterostructure that enables such functionality, using a magneto-optical material embedded between two dissimilar dielectric layers., Comment: 9 pages, 6 figures

The Pyramid Wavefront Sensor (PyWFS) is highly nonlinear and requires the use of beam modulation to successfully close an AO loop under varying atmospheric turbulence conditions, at the expense of a loss in sensitivity. In this work we train, analyse, and compare the use of deep neural networks (NNs) as non-linear estimators for the non-modulated PyWFS, identifying the most suitable NN architecture for reliable closed-loop AO. We develop a novel training strategy for NNs that seeks to accommodate for changes in residual statistics between open and closed-loop, plus the addition of noise for robustness purposes. Through simulations, we test and compare several deep NNs, from classical to new convolutional neural networks (CNNs), plus a state-of-the-art transformer neural network (TNN, Global Context Visual Transformer, GCViT), first in open-loop and then in closed-loop. Using open-loop simulated data, we observe that a TNN (GCViT) largely surpasses any CNN in estimation accuracy in a wide range of turbulence conditions. Also, the TNN performs better in simulated closed-loop than CNNs, avoiding estimation issues at the pupil borders. When closing the loop at strong turbulence and low noise, the TNN using non-modulated PyWFS data is able to close the loop similar to a PyWFS with $12\lambda/D$ of modulation. When raising the noise only the TNN is able to close the loop, while the standard linear reconstructor fails, even with modulation. Using the GCViT, we close a real AO loop in the optical bench achieving a Strehl ratio between 0.28 and 0.77 for turbulence conditions ranging from 6cm to 20cm, respectively. In conclusion, we demonstrate that a TNN is the most suitable architecture to extend the dynamic range without sacrificing sensitivity for a non-modulated PyWFS. It opens the path for using non-modulated Pyramid WFSs under an unprecedented range of atmospheric and noise conditions., Comment: 8 pages, 9 figures

The increasing sophistication of cyber threats necessitates proactive measures to identify vulnerabilities and potential exploits. Underground hacking forums serve as breeding grounds for the exchange of hacking techniques and discussions related to exploitation. In this research, we propose an innovative approach using topic modeling to analyze and uncover key themes in vulnerabilities discussed within these forums. The objective of our study is to develop a machine learning-based model that can automatically detect and classify vulnerability-related discussions in underground hacking forums. By monitoring and analyzing the content of these forums, we aim to identify emerging vulnerabilities, exploit techniques, and potential threat actors. To achieve this, we collect a large-scale dataset consisting of posts and threads from multiple underground forums. We preprocess and clean the data to ensure accuracy and reliability. Leveraging topic modeling techniques, specifically Latent Dirichlet Allocation (LDA), we uncover latent topics and their associated keywords within the dataset. This enables us to identify recurring themes and prevalent discussions related to vulnerabilities, exploits, and potential targets., Comment: 6 pages

In 2021, the pioneering work on TypeNet showed that keystroke dynamics verification could scale to hundreds of thousands of users with minimal performance degradation. Recently, the KVC-onGoing competition has provided an open and robust experimental protocol for evaluating keystroke dynamics verification systems of such scale, including considerations of algorithmic fairness. This article describes Type2Branch, the model and techniques that achieved the lowest error rates at the KVC-onGoing, in both desktop and mobile scenarios. The novelty aspects of the proposed Type2Branch include: i) synthesized timing features emphasizing user behavior deviation from the general population, ii) a dual-branch architecture combining recurrent and convolutional paths with various attention mechanisms, iii) a new loss function named Set2set that captures the global structure of the embedding space, and iv) a training curriculum of increasing difficulty. Considering five enrollment samples per subject of approximately 50 characters typed, the proposed Type2Branch achieves state-of-the-art performance with mean per-subject EERs of 0.77% and 1.03% on evaluation sets of respectively 15,000 and 5,000 subjects for desktop and mobile scenarios. With a uniform global threshold for all subjects, the EERs are 3.25% for desktop and 3.61% for mobile, outperforming previous approaches by a significant margin., Comment: 13 pages, 3 figures

For long document summarization, discourse structure is important to discern the key content of the text and the differences in importance level between sentences. Unfortunately, the integration of rhetorical structure theory (RST) into parameter-efficient fine-tuning strategies for long document summarization remains unexplored. Therefore, this paper introduces RST-LoRA and proposes four RST-aware variants to explicitly incorporate RST into the LoRA model. Our empirical evaluation demonstrates that incorporating the type and uncertainty of rhetorical relations can complementarily enhance the performance of LoRA in summarization tasks. Furthermore, the best-performing variant we introduced outperforms the vanilla LoRA and full-parameter fine-tuning models, as confirmed by multiple automatic and human evaluations, and even surpasses previous state-of-the-art methods., Comment: NAACL 2024 Main & Long Conference Paper (Oral Presentation)

Widely deployed large language models (LLMs) can produce convincing yet incorrect outputs, potentially misleading users who may rely on them as if they were correct. To reduce such overreliance, there have been calls for LLMs to communicate their uncertainty to end users. However, there has been little empirical work examining how users perceive and act upon LLMs' expressions of uncertainty. We explore this question through a large-scale, pre-registered, human-subject experiment (N=404) in which participants answer medical questions with or without access to responses from a fictional LLM-infused search engine. Using both behavioral and self-reported measures, we examine how different natural language expressions of uncertainty impact participants' reliance, trust, and overall task performance. We find that first-person expressions (e.g., "I'm not sure, but...") decrease participants' confidence in the system and tendency to agree with the system's answers, while increasing participants' accuracy. An exploratory analysis suggests that this increase can be attributed to reduced (but not fully eliminated) overreliance on incorrect answers. While we observe similar effects for uncertainty expressed from a general perspective (e.g., "It's not clear, but..."), these effects are weaker and not statistically significant. Our findings suggest that using natural language expressions of uncertainty may be an effective approach for reducing overreliance on LLMs, but that the precise language used matters. This highlights the importance of user testing before deploying LLMs at scale., Comment: Accepted to FAccT 2024. This version includes the appendix

Cancer cells exhibit significant alterations in their metabolism, characterised by a reduction in oxidative phosphorylation (OXPHOS) and an increased reliance on glycolysis, even in the presence of oxygen. This metabolic shift, known as the Warburg effect, is pivotal in fuelling cancer's uncontrolled growth, invasion, and therapeutic resistance. While dysregulation of many genes contributes to this metabolic shift, the tumour suppressor gene p53 emerges as a master player. Yet, the molecular mechanisms remain elusive. This study introduces a comprehensive mathematical model, integrating essential p53 targets, offering insights into how p53 orchestrates its targets to redirect cancer metabolism towards an OXPHOS-dominant state. Simulation outcomes align closely with experimental data comparing glucose metabolism in colon cancer cells with wild-type and mutated p53. Additionally, our findings reveal the dynamic capability of elevated p53 activation to fully reverse the Warburg effect, highlighting the significance of its activity levels not just in triggering apoptosis (programmed cell death) post-chemotherapy but also in modifying the metabolic pathways implicated in treatment resistance. In scenarios of p53 mutations, our analysis suggests targeting glycolysis-instigating signalling pathways as an alternative strategy, whereas targeting solely synthesis of cytochrome c oxidase 2 (SCO2) does support mitochondrial respiration but may not effectively suppress the glycolysis pathway, potentially boosting the energy production and cancer cell viability.

Nigerian Pidgin is an English-derived contact language and is traditionally an oral language, spoken by approximately 100 million people. No orthographic standard has yet been adopted, and thus the few available Pidgin datasets that exist are characterised by noise in the form of orthographic variations. This contributes to under-performance of models in critical NLP tasks. The current work is the first to describe various types of orthographic variations commonly found in Nigerian Pidgin texts, and model this orthographic variation. The variations identified in the dataset form the basis of a phonetic-theoretic framework for word editing, which is used to generate orthographic variations to augment training data. We test the effect of this data augmentation on two critical NLP tasks: machine translation and sentiment analysis. The proposed variation generation framework augments the training data with new orthographic variants which are relevant for the test set but did not occur in the training set originally. Our results demonstrate the positive effect of augmenting the training data with a combination of real texts from other corpora as well as synthesized orthographic variation, resulting in performance improvements of 2.1 points in sentiment analysis and 1.4 BLEU points in translation to English., Comment: Accepted to LREC-COLING 2024 Main Conference

We have determined the depth-dose curve, the penetration range, and the water equivalent ratio (WER), for proton beams of clinical energies in cortical bone, by means of a detailed and accurate simulation that combines molecular dynamics and Monte Carlo techniques. The fundamental input quantities (stopping power and energy loss straggling) for the simulation were obtained from a reliable electronic excitation spectrum of the condensed-phase target, which takes into account the organic and mineral phases that form it. Our simulations with these inputs, that are in excellent agreement with the scarce data available for a cortical bone target, deviate from simulations performed using other stopping quantities, such as those provided by the International Commission on Radiation Units and Measurements (ICRU) in its widely used Report 49. The results of this work emphasize the importance of an accurate determination of the stopping quantities of cortical bone in order to advance towards the millimetric precision for the proton penetration ranges and deposited dose needed in radiotherapy., Comment: Accepted version in Physical Review E (see published version in: https://journals.aps.org/pre/abstract/10.1103/PhysRevE.110.034405), 20 pages, 6 figures

The tidal problem is used to obtain the tidal deformability (or Love number) of stars. The semi-analytical study is usually treated in perturbation theory as a first order perturbation problem over a spherically symmetric background configuration consisting of a stellar interior region matched across a boundary to a vacuum exterior region that models the tidal field. The field equations for the metric and matter perturbations at the interior and exterior regions are complemented with corresponding boundary conditions. The data of the two problems at the common boundary are related by the so called matching conditions. These conditions for the tidal problem are known in the contexts of perfect fluid stars and superfluid stars modelled by a two-fluid. Here we review the obtaining of the matching conditions for the tidal problem starting from a purely geometrical setting, and present them so that they can be readily applied to more general contexts, such as other types of matter fields, different multiple layers or phase transitions. As a guide on how to use the matching conditions, we recover the known results for perfect fluid and superfluid neutron stars.

The X-ray Integral Field Unit (X-IFU) is the high-resolution X-ray spectrometer to fly on board the Athena Space Observatory of the European Space Agency (ESA). It is being developed by an international Consortium led by France, involving twelve ESA member states, plus the United States. It is a cryogenic instrument, involving state of the art technology, such as micro-calorimeters, to be read out by low noise electronics. As the instrument was undergoing its system requirement review (in 2022), a life cycle assessment (LCA) was performed to estimate the environmental impacts associated with the development of the sub-systems that were under the responsibility of the X-IFU Consortium. The assessment included the supply, manufacturing and testing of sub systems, as well as involved logistics and manpower. We find that the most significant environmental impacts arise from testing activities, which is related to energy consumption in clean rooms, office work, which is related to energy consumption in office buildings, and instrument manufacturing, which is related to the use of mineral and metal resources. Furthermore, business travels is another area of concern, despite the policy to reduced flying adopted by the Consortium. As the instrument is now being redesigned to fit within the new boundaries set by ESA, the LCA will be updated, with a focus on the hot spots identified in the first iteration. The new configuration, consolidated in 2023, is significantly different from the previously studied version and is marked by an increase of the perimeter of responsibility for the Consortium. This will need to be folded in the updated LCA, keeping the ambition to reduce the environmental footprint of X-IFU, while complying with its stringent requirements in terms of performance and risk management., Comment: 28 pages, 7 figures, accepted for publication in Experimental Astronomy

In this paper, we propose the replication of scientific experiments, with the participation of the general public, as one of the possible strategies to confront science deniers. Using the social media of the Federal University of Lavras (UFLA), the project entitled : The Magic of Physics and the Universe, invited the public to reproduce the experiment carried out by the Greek Eratosthenes over 2,100 years ago, which he proved that the Earth is spherical. We present a report of this experience to suggest that it is possible for citizens to experience scientific knowledge and observe the phenomena in a simple way, in order to develop a critical interpretation of false information that circulates in society. This proposal distinctively geeks to deconstruct the denial of the Earth's curvature through direct and collective participation of citizens in Eratosthenes' experiment, which permit more reflection on how the scientific method used to study this subject works., Comment: This article is written in Portuguese

Introduction and Objective. Assuming the need to rethink Education and its function as an element of social transformation and generator of a profound ethical change and promoter of other ways of being on planet Earth, this study seeks to understand the possible dialogues between Earth Pedagogy and Astrobiology in the construction and proposition of continuing education processes for science teachers. Methodology. A mini-course was promoted aimed at teachers and an attempt was made to report the construction process of the training process and, using Content Analysis, to analyze the contributions pointed out by the participants to their training and teaching practice. Results and Conclusion. The dialogue between knowledge in the continuing education of teachers contributes with fundamental elements to teaching practice, covering cognitive, emotional, individual and collective aspects, as well as highlighting the role of the university in promoting initiatives that bring Education and Complexity together., Comment: This article is written in Portuguese

We introduce Sylow subgroups and $0$-groups to the theory of complex algebraic supergroups, which mimic Sylow subgroups and $p$-groups in the theory of finite groups. We prove that Sylow subgroups are always $0$-groups, and show that they are unique up to conjugacy. Further, we give an explicit classification of $0$-groups which will be very useful for future applications. Finally, we prove an analogue of Sylow's third theorem on the number of Sylow subgroups of a supergroup., Comment: 33 pages; comments welcome!

Synthetic data is gaining increasing relevance for training machine learning models. This is mainly motivated due to several factors such as the lack of real data and intra-class variability, time and errors produced in manual labeling, and in some cases privacy concerns, among others. This paper presents an overview of the 2nd edition of the Face Recognition Challenge in the Era of Synthetic Data (FRCSyn) organized at CVPR 2024. FRCSyn aims to investigate the use of synthetic data in face recognition to address current technological limitations, including data privacy concerns, demographic biases, generalization to novel scenarios, and performance constraints in challenging situations such as aging, pose variations, and occlusions. Unlike the 1st edition, in which synthetic data from DCFace and GANDiffFace methods was only allowed to train face recognition systems, in this 2nd edition we propose new sub-tasks that allow participants to explore novel face generative methods. The outcomes of the 2nd FRCSyn Challenge, along with the proposed experimental protocol and benchmarking contribute significantly to the application of synthetic data to face recognition., Comment: arXiv admin note: text overlap with arXiv:2311.10476

In this paper we consider the problem of acoustic inversion in the context of the optoacoustic tomography image reconstruction problem. By leveraging the ability of the recently proposed diffusion models for image generative tasks among others, we devise an image reconstruction architecture based on a conditional diffusion process. The scheme makes use of an initial image reconstruction, which is preprocessed by an autoencoder to generate an adequate representation. This representation is used as conditional information in a generative diffusion process. Although the computational requirements for training and implementing the architecture are not low, several design choices discussed in the work were made to keep them manageable. Numerical results show that the conditional information allows to properly bias the parameters of the diffusion model to improve the quality of the initial reconstructed image, eliminating artifacts or even reconstructing finer details of the ground-truth image that are not recoverable by the initial image reconstruction method. We also tested the proposal under experimental conditions and the obtained results were in line with those corresponding to the numerical simulations. Improvements in image quality up to 17 % in terms of peak signal-to-noise ratio were observed., Comment: Paper accepted for publication in the journal Optics and Lasers in Engineering

This article provides a comprehensive overview of recent research in the area of Child-Computer Interaction (CCI). The main contributions of the present article are two-fold. First, we present a novel longitudinal CCI database named ChildCIdbLong, which comprises over 600 children aged 18 months to 8 years old, acquired continuously over 4 academic years (2019-2023). As a result, ChildCIdbLong comprises over 12K test acquisitions over a tablet device. Different tests are considered in ChildCIdbLong, requiring different touch and stylus gestures, enabling the evaluation of praxical and cognitive skills such as attentional, visuo-spatial, and executive, among others. In addition to the ChildCIdbLong database, we propose a novel quantitative metric called Test Quality (Q), designed to measure the motor and cognitive development of children through their interaction with a tablet device. In order to provide a better comprehension of the proposed Q metric, popular percentile-based growth representations are introduced for each test, providing a two-dimensional space to compare children's development with respect to the typical age skills of the population. The results achieved in the present article highlight the potential of the novel ChildCIdbLong database in conjunction with the proposed Q metric to measure the motor and cognitive development of children as they grow up. The proposed framework could be very useful as an automatic tool to support child experts (e.g., paediatricians, educators, or neurologists) for early detection of potential physical/cognitive impairments during children's development., Comment: 13 pages, 5 figures, 7 tables, 46 references

H$_2$CO is a small organic molecule widely detected in protoplanetary disks. As a precursor to grain-surface formation of CH$_3$OH, H$_2$CO is considered an important precursor of O-bearing organic molecules that are locked in ices. Still, since gas-phase reactions can also form H$_2$CO, there remains an open question on the channels by which organics form in disks, and how much the grain versus the gas pathways impact the overall organic reservoir. We present spectrally and spatially resolved Atacama Large Millimeter/submillimeter Array observations of several ortho- and para-H$_2$CO transitions toward the bright protoplanetary disk around the Herbig Ae star HD 163296. We derive column density, excitation temperature, and ortho-to-para ratio (OPR) radial profiles for H$_2$CO, as well as disk-averaged values of $N_{\mathrm{T}}\sim4\times 10^{12}$ cm$^{-2}$, $T_{\mathrm{ex}}\sim20$ K, and $\mathrm{OPR}\sim2.7$, respectively. We empirically determine the vertical structure of the emission, finding vertical heights of $z/r\sim0.1$. From the profiles, we find a relatively constant $\mathrm{OPR}\sim2.7$ with radius, but still consistent with $3.0$ among the uncertainties, a secondary increase of $N_{\mathrm{T}}$ in the outer disk, and low $T_{\mathrm{ex}}$ values that decrease with disk radius. Our resulting radial, vertical, and OPR constraints suggest an increased UV penetration beyond the dust millimeter edge, consistent with an icy origin but also with cold gas-phase chemistry. This Herbig disk contrasts previous results for the T Tauri disk, TW Hya, which had a larger contribution from cold gas-phase chemistry. More observations of other sources are needed to disentangle the dominant formation pathway of H$_2$CO in protoplanetary disks., Comment: 21 pages, 4 tables, 10 figures; V2: minor changes, matches the published version in ApJ

Large-scale face recognition datasets are collected by crawling the Internet and without individuals' consent, raising legal, ethical, and privacy concerns. With the recent advances in generative models, recently several works proposed generating synthetic face recognition datasets to mitigate concerns in web-crawled face recognition datasets. This paper presents the summary of the Synthetic Data for Face Recognition (SDFR) Competition held in conjunction with the 18th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2024) and established to investigate the use of synthetic data for training face recognition models. The SDFR competition was split into two tasks, allowing participants to train face recognition systems using new synthetic datasets and/or existing ones. In the first task, the face recognition backbone was fixed and the dataset size was limited, while the second task provided almost complete freedom on the model backbone, the dataset, and the training pipeline. The submitted models were trained on existing and also new synthetic datasets and used clever methods to improve training with synthetic data. The submissions were evaluated and ranked on a diverse set of seven benchmarking datasets. The paper gives an overview of the submitted face recognition models and reports achieved performance compared to baseline models trained on real and synthetic datasets. Furthermore, the evaluation of submissions is extended to bias assessment across different demography groups. Lastly, an outlook on the current state of the research in training face recognition models using synthetic data is presented, and existing problems as well as potential future directions are also discussed., Comment: The 18th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2024)

The Schm\"udgen's Positivstellensatz gives a certificate to verify positivity of a strictly positive polynomial $f$ on a compact, basic, semi-algebraic set $\mathbf{K} \subset \mathbb{R}^n$. A Positivstellensatz of this type is called effective if one may bound the degrees of the polynomials appearing in the certificate in terms of properties of $f$. If $\mathbf{K} = [-1,1]^n$ and $0 < f_\min := \min_{x \in \mathbf{K}} f(x)$, then the degrees of the polynomials appearing in the certificate may be bounded by $O\left(\sqrt{\frac{f_\max - f_\min}{f_\min}}\right)$, where $f_\max := \max_{x \in \mathbf{K}} f(x)$, as was recently shown by Laurent and Slot [Optimization Letters 17:515-530, 2023]. The big-O notation suppresses dependence on $n$ and the degree $d$ of $f$. In this paper we show a similar result, but with a better dependence on $n$ and $d$. In particular, our bounds depend on the $1$-norm of the coefficients of $f$, that may readily be calculated.

In this paper, we examine the stability of a heat-conducting copper rod and a magnetizable piezoelectric beam, with fractional damping affecting the longitudinal displacement of the piezoelectric material's centerline. We establish a polynomial stability result that is dependent on the order of the fractional derivative., Comment: arXiv admin note: text overlap with arXiv:2204.00283, arXiv:2208.12012, arXiv:2102.07538 by other authors

We establish two main results for the asymptotic dimension of countable approximate groups. The first one is a Hurewicz type formula for a global morphism of countable approximate groups $f:(\Xi, \Xi^\infty) \to (\Lambda, \Lambda^\infty)$, stating that $\mathrm{asdim} \Xi \leq \mathrm{asdim} \Lambda +\mathrm{asdim} ([\mathrm{ker} f]_c)$. This is analogous to the Dranishnikov-Smith result for groups, and is relying on another Hurewicz type formula we prove, using a 6-local morphism instead of a global one. The second result is similar to the Dranishnikov-Smith theorem stating that, for a countable group $G$, $\mathrm{asdim} G$ is equal to the supremum of asymptotic dimensions of finitely generated subgroups of $G$. Our version states that, if $(\Lambda, \Lambda^\infty)$ is a countable approximate group, then $\mathrm{asdim} \Lambda$ is equal to the supremum of asymptotic dimensions of approximate subgroups of finitely generated subgroups of $\Lambda^\infty$, with these approximate subgroups contained in $\Lambda^2$., Comment: The results in this paper were previously contained in the monograph titled "Foundations of geometric approximate group theory", by M. Cordes and the two authors listed here, but they had to be taken out of that monograph in order to shorten it. arXiv admin note: substantial text overlap with arXiv:2012.15303

Stochastic battery bidding in real-time energy markets is a nuanced process, with its efficacy depending on the accuracy of forecasts and the representative scenarios chosen for optimization. In this paper, we introduce a pioneering methodology that amalgamates Transformer-based forecasting with weighted constrained Dynamic Time Warping (wcDTW) to refine scenario selection. Our approach harnesses the predictive capabilities of Transformers to foresee Energy prices, while wcDTW ensures the selection of pertinent historical scenarios by maintaining the coherence between multiple uncertain products. Through extensive simulations in the PJM market for July 2023, our method exhibited a 10% increase in revenue compared to the conventional method, highlighting its potential to revolutionize battery bidding strategies in real-time markets.