Search

We present a full one-loop calculation of the gravitino thermal production rate, beyond the so-called hard thermal loop approximation, using the corresponding thermal spectral functions in numerical form on both sides of the light cone. This framework requires a full numerical evaluation. We interpret our results within the framework of a general supergravity-based model, remaining agnostic about the specifics of supersymmetry breaking. In this context, assuming that gravitinos constitute the entirety of the dark matter in the Universe imposes strict constraints on the reheating temperature. For example, with a gluino mass at the current LHC limit, a maximum reheating temperature of $T_\mathrm{reh} \simeq 10^9$ GeV is compatible with a gravitino mass of $m_{3/2} \simeq 1$ TeV. Additionally, with a reheating temperature an order of magnitude lower at $T_\mathrm{reh} \simeq 10^8$ GeV, the common gaugino mass $M_{1/2}$ can range from $2$ to $4 $ TeV within the same gravitino mass range. For much higher values of $M_{1/2}$, which are favored by current accelerator and cosmological data in the context of supersymmetric models, such as $M_{1/2} = 10$ TeV, and for $m_{3/2} \simeq 1$ TeV the reheating temperature compatible with the gravitino dark matter scenario is $ 10^7$ GeV. If other dark matter particles are considered, the reheating temperature could be much lower., Comment: 53 pages, 15 figures

Consider the $d$-dimensional hyperbolic space $\mathbb{M}_K^d$ of constant curvature $K<0$ and fix a point $o$ playing the role of an origin. Let $\mathbf{L}$ be a uniform random $q$-dimensional totally geodesic submanifold (called $q$-flat) in $\mathbb{M}_K^d$ passing through $o$ and, independently of $\mathbf{L}$, let $\mathbf{E}$ be a random $(d-q+\gamma)$-flat in $\mathbb{M}_K^d$ which is uniformly distributed in the set of all $(d-q+\gamma)$-flats intersecting a hyperbolic ball of radius $u>0$ around $o$. We are interested in the distribution of the random $\gamma$-flat arising as the intersection of $\mathbf{E}$ with $\mathbf{L}$. In contrast to the Euclidean case, the intersection $\mathbf{E}\cap \mathbf{L}$ can be empty with strictly positive probability. We determine this probability and the full distribution of $\mathbf{E}\cap \mathbf{L}$. Thereby, we elucidate crucial differences to the Euclidean case. Moreover, we study the limiting behaviour as $d\uparrow\infty$ and also $K\uparrow 0$. Thereby we obtain a phase transition with three different phases which we completely characterize, including a critical phase with distinctive behavior and a phase recovering the Euclidean results. In the background are methods from hyperbolic integral geometry.

We study the anomalous magnetic moment of the muon, $g_\mu - 2 \equiv 2 a_\mu$, in the context of supersymmetric models beyond the CMSSM, where the unification of either the gaugino masses $M_{1,2,3}$ or sfermion and Higgs masses is relaxed, taking into account the measured mass of the Higgs boson, $m_H$, the cosmological dark matter density and the direct detection rate. We find that the model with non-unified gaugino masses can make a contribution $\Delta a_\mu\sim 20 \times 10^{-10}$ to the anomalous magnetic moment of the muon, for example if $M_{1,2} \sim 600$ GeV and $M_3\sim 8$ TeV. The model with non-universal sfermion and Higgs masses can provide even larger $\Delta a_\mu \sim 24 \times 10^{-10}$ if the sfermion masses for the first and the second generations are $ \sim 400 $ GeV and that of the third is $ \sim 8 $ TeV. We discuss the prospects for collider searches for supersymmetric particles in specific benchmark scenarios illustrating these possibilities, focusing in particular on the prospects for detecting the lighter smuon and the lightest neutralino., Comment: 34 pages, 17 figures

The absence of well-structured large datasets in medical computer vision results in decreased performance of automated systems and, especially, of deep learning models. Domain generalization techniques aim to approach unknown domains from a single data source. In this paper we introduce a novel framework, named CompStyle, which leverages style transfer and adversarial training, along with high-level input complexity augmentation to effectively expand the domain space and address unknown distributions. State-of-the-art style transfer methods depend on the existence of subdomains within the source dataset. However, this can lead to an inherent dataset bias in the image creation. Input-level augmentation can provide a solution to this problem by widening the domain space in the source dataset and boost performance on out-of-domain distributions. We provide results from experiments on semantic segmentation on prostate data and corruption robustness on cardiac data which demonstrate the effectiveness of our approach. Our method increases performance in both tasks, without added cost to training time or resources., Comment: Accepted at IEEE/CVF Computer Vision and Pattern Recognition Conference Workshops (CVPRW) 2024

Safe reinforcement learning (RL) is crucial for deploying RL agents in real-world applications, as it aims to maximize long-term rewards while satisfying safety constraints. However, safe RL often suffers from sample inefficiency, requiring extensive interactions with the environment to learn a safe policy. We propose Efficient Safe Policy Optimization (ESPO), a novel approach that enhances the efficiency of safe RL through sample manipulation. ESPO employs an optimization framework with three modes: maximizing rewards, minimizing costs, and balancing the trade-off between the two. By dynamically adjusting the sampling process based on the observed conflict between reward and safety gradients, ESPO theoretically guarantees convergence, optimization stability, and improved sample complexity bounds. Experiments on the Safety-MuJoCo and Omnisafe benchmarks demonstrate that ESPO significantly outperforms existing primal-based and primal-dual-based baselines in terms of reward maximization and constraint satisfaction. Moreover, ESPO achieves substantial gains in sample efficiency, requiring 25--29% fewer samples than baselines, and reduces training time by 21--38%.

Assistive Exoskeleton Robots are helping restore functions to people suffering from underlying medical conditions. These robots require precise tuning of hyper-parameters to feel natural to the user. The device hyper-parameters often need to be re-tuned from task to task, which can be tedious and require expert knowledge. To address this issue, we develop a set of fuzzy logic controllers that can dynamically tune robot gain parameters to adapt its sensitivity to the user's intention determined from muscle activation. The designed fuzzy controllers benefit from a set of expert-defined rules and do not rely on extensive amounts of training data. We evaluate the designed controllers with three different tasks and compare our results against the manually tuned system. Our preliminary results show that our controllers reduce the amount of fighting between the device and the human, measured using a set of pressure sensors., Comment: 4 pages, 5 figures, Accepted to 2024 21st International Conference on Ubiquitous Robots (UR)

In the domain of assistive robotics, the significance of effective modeling is well acknowledged. Prior research has primarily focused on enhancing model accuracy or involved the collection of extensive, often impractical amounts of data. While improving individual model accuracy is beneficial, it necessitates constant remodeling for each new task and user interaction. In this paper, we investigate the generalizability of different modeling methods. We focus on constructing the dynamic model of an assistive exoskeleton using six data-driven regression algorithms. Six tasks are considered in our experiments, including horizontal, vertical, diagonal from left leg to the right eye and the opposite, as well as eating and pushing. We constructed thirty-six unique models applying different regression methods to data gathered from each task. Each trained model's performance was evaluated in a cross-validation scenario, utilizing five folds for each dataset. These trained models are then tested on the other tasks that the model is not trained with. Finally the models in our study are assessed in terms of generalizability. Results show the superior generalizability of the task model performed along the horizontal plane, and decision tree based algorithms., Comment: Accepted to 2024 21st International Conference on Ubiquitous Robots (UR)

Let $G$ be a vertex-transitive graph of superlinear polynomial growth. Given $r>0$, let $G_r$ be the graph on the same vertex set as $G$, with two vertices joined by an edge if and only if they are at graph distance at most $r$ apart in $G$. We show that the critical probability $p_c(G_r)$ for Bernoulli bond percolation on $G_r$ satisfies $p_c(G_r) \sim 1/\mathrm{deg}(G_r)$ as $r\to\infty$. This extends work of Penrose and Bollob\'as-Janson-Riordan, who considered the case $G=\mathbb{Z}^d$. Our result provides an important ingredient in parallel work of Georgakopoulos in which he introduces a new notion of dimension in groups. It also verifies a special case of a conjecture of Easo and Hutchcroft., Comment: 35 pages

Reinforcement learning (RL) is a powerful tool for optimal control that has found great success in Atari games, the game of Go, robotic control, and building optimization. RL is also very brittle; agents often overfit to their training environment and fail to generalize to new settings. Unsupervised environment design (UED) has been proposed as a solution to this problem, in which the agent trains in environments that have been specially selected to help it learn. Previous UED algorithms focus on trying to train an RL agent that generalizes across a large distribution of environments. This is not necessarily desirable when we wish to prioritize performance in one environment over others. In this work, we will be examining the setting of robust RL building control, where we wish to train an RL agent that prioritizes performing well in normal weather while still being robust to extreme weather conditions. We demonstrate a novel UED algorithm, ActivePLR, that uses uncertainty-aware neural network architectures to generate new training environments at the limit of the RL agent's ability while being able to prioritize performance in a desired base environment. We show that ActivePLR is able to outperform state-of-the-art UED algorithms in minimizing energy usage while maximizing occupant comfort in the setting of building control.

Financial portfolio management describes the task of distributing funds and conducting trading operations on a set of financial assets, such as stocks, index funds, foreign exchange or cryptocurrencies, aiming to maximize the profit while minimizing the loss incurred by said operations. Deep Learning (DL) methods have been consistently excelling at various tasks and automated financial trading is one of the most complex one of those. This paper aims to provide insight into various DL methods for financial trading, under both the supervised and reinforcement learning schemes. At the same time, taking into consideration sentiment information regarding the traded assets, we discuss and demonstrate their usefulness through corresponding research studies. Finally, we discuss commonly found problems in training such financial agents and equip the reader with the necessary knowledge to avoid these problems and apply the discussed methods in practice.

The development of an electronic voting system that would replace traditional election procedures is a research topic of great interest for many years. Blockchain technology could provide some guarantees and fulfill strong requirements for electronic voting platforms, such as transparency, immutability, and confidentiality. From time to time research is conducted to address problems in voting systems. Many research works attempt to implement secure and reliable voting systems, which address known security, anonymity, and fraud issues that might threaten such systems. This paper presents a proposal of a secure electronic voting system, the EtherVote, using the Ethereum Blockchain network that focuses deeply on the field of identification of eligible citizens. The proposed system will be entirely based on Blockchain without any central authority servers or databases, thus improving security, privacy, and election cost. Limitations, problems, and solutions are discussed, in order to make the proposed electronic voting system ideal and ready to use for national elections., Comment: 2 pages, Poster presented in ACM 5th summit on Gender Equality in Computing, GEC 2023, Athens University of Economics and Business, Athens, Greece, 27 June 2023

The electronic design industry has undergone a significant transformation, transitioning from traditional hand-drawn designs to modern automated design processes. While Computer-Aided Design (CAD) tools emerged alongside the electronic industry, the current building design process has little to no automation. There is a need for a unified platform to address the complexity of building design and provide a more systematic approach. Platform-based design (PBD), originally developed in the electronic industry, enables efficient design processes by promoting the reuse of hardware and software systems. It also facilitates design space exploration while optimizing performance. This paper proposes a modular approach that divides the building into various disciplines and introduces a design flow using the PBD framework to streamline the design process. We also present a case study that demonstrates the use of the PBD framework in the Heating, Ventilation, and Air Conditioning (HVAC) systems design., Comment: 17 pages, 3 figures

Design automation, which involves the use of software tools and technologies to streamline the design process, has been widely adopted in the electronics industry, resulting in significant advancements in product development and manufacturing. However, building design, which involves the creation of complex structures and systems, has traditionally lagged behind in leveraging design automation technologies. Despite extensive research on design automation in the building industry, its application in the current design of buildings is limited. This paper aims to (1) compare the design processes between electronics and building design, (2) highlight similarities and differences in their approaches, and (3) examine challenges and opportunities associated with bringing the concept of design automation from electronics to building design.

The freeze-in mechanism has been shown to allow the simultaneous generation of cosmic dark matter and a viable matter-antimatter asymmetry in the universe. When the underlying interactions are described by higher-dimensional, non-renormalizable operators, the relevant freeze-in processes take place close to the highest considered cosmic temperatures. In this paper we study how the presence of a fluid that temporarily dominates the energy content of the early universe affects the predictions of this ``Ultraviolet Freeze-In Baryogenesis'' scenario. We find that this additional cosmic component has a significant impact on the predictions of concrete microscopic models, allowing for reheating temperatures which are much lower than those required in the simplest cosmological scenario. Moreover, we show that inflationary observables can constrain the parameter space of such models, once the latter are examined in conjunction with concrete models of inflation., Comment: 25 pages, 6 figures

Optical activity (polarization rotation of light) is one of the most desired features of chiral media, as it is important for many polarization related applications. However, in the THz region, chiral media with strong optical activity are not available in nature. Here, we study theoretically, and experimentally a chiral metamaterial structure composed of pairs of vertical U-shape resonators of "twisted" arms, and we reveal that it demonstrates large pure optical activity (i.e. optical activity associated with negligible transmitted wave ellipticity) in the low THz regime. The experimental data show polarization rotation up to 25 (deg) for an unmatched bandwidth of 1 THz (relative bandwidth 80 %), from a 130 um-thickness structure, while theoretical optimizations show that the rotation can reach 45 (deg). The enhanced chiral response of the structure is analyzed through an equivalent RLC circuit model, which provides also simple optimization rules for the enhancement of its chiral response. The proposed chiral structures allow easy fabrication via direct laser writing and electroless metal plating, making them suitable candidates for polarization control applications., Comment: 17 pages, 7 figures

The Biden Administration expects to resettle 95,000 Afghan evacuees in communities across the United States by September 2022. Some evacuees come to the United States with strong English proficiency and years of experience working as entrepreneurs or community leaders or for the United States military. Others, especially women, have had limited opportunities to achieve an education or learn English. City governments and resettlement agencies cannot respond to these diverse needs on their own. Because of this, the State Department launched a partnership with the nonprofit Welcome.US to coordinate the resettlement efforts to meet these disparate needs and encourage groups working with Afghan evacuees to think creatively about how they can channel their unique skills and resources to help our new Afghan neighbors.

Energy consumption in buildings, both residential and commercial, accounts for approximately 40% of all energy usage in the U.S., and similar numbers are being reported from countries around the world. This significant amount of energy is used to maintain a comfortable, secure, and productive environment for the occupants. So, it is crucial that the energy consumption in buildings must be optimized, all the while maintaining satisfactory levels of occupant comfort, health, and safety. Recently, Machine Learning has been proven to be an invaluable tool in deriving important insights from data and optimizing various systems. In this work, we review the ways in which machine learning has been leveraged to make buildings smart and energy-efficient. For the convenience of readers, we provide a brief introduction of several machine learning paradigms and the components and functioning of each smart building system we cover. Finally, we discuss challenges faced while implementing machine learning algorithms in smart buildings and provide future avenues for research at the intersection of smart buildings and machine learning.