Search

Due to the rapid development of science and technology, the importance of imprecise, noisy, and uncertain data is increasing at an exponential rate. Thus, mining patterns in uncertain databases have drawn the attention of researchers. Moreover, frequent sequences of items from these databases need to be discovered for meaningful knowledge with great impact. In many real cases, weights of items and patterns are introduced to find interesting sequences as a measure of importance. Hence, a constraint of weight needs to be handled while mining sequential patterns. Besides, due to the dynamic nature of databases, mining important information has become more challenging. Instead of mining patterns from scratch after each increment, incremental mining algorithms utilize previously mined information to update the result immediately. Several algorithms exist to mine frequent patterns and weighted sequences from incremental databases. However, these algorithms are confined to mine the precise ones. Therefore, we have developed an algorithm to mine frequent sequences in an uncertain database in this work. Furthermore, we have proposed two new techniques for mining when the database is incremental. Extensive experiments have been conducted for performance evaluation. The analysis showed the efficiency of our proposed framework., Comment: Accepted to Information Science journal

In this uncertain world, data uncertainty is inherent in many applications and its importance is growing drastically due to the rapid development of modern technologies. Nowadays, researchers have paid more attention to mine patterns in uncertain databases. A few recent works attempt to mine frequent uncertain sequential patterns. Despite their success, they are incompetent to reduce the number of false-positive pattern generation in their mining process and maintain the patterns efficiently. In this paper, we propose multiple theoretically tightened pruning upper bounds that remarkably reduce the mining space. A novel hierarchical structure is introduced to maintain the patterns in a space-efficient way. Afterward, we develop a versatile framework for mining uncertain sequential patterns that can effectively handle weight constraints as well. Besides, with the advent of incremental uncertain databases, existing works are not scalable. There exist several incremental sequential pattern mining algorithms, but they are limited to mine in precise databases. Therefore, we propose a new technique to adapt our framework to mine patterns when the database is incremental. Finally, we conduct extensive experiments on several real-life datasets and show the efficacy of our framework in different applications., Comment: Accepted at PAKDD 2021. arXiv admin note: text overlap with arXiv:2404.00746

In the dynamic and ever-changing domain of Unmanned Aerial Vehicles (UAVs), the utmost importance lies in guaranteeing resilient and lucid security measures. This study highlights the necessity of implementing a Zero Trust Architecture (ZTA) to enhance the security of unmanned aerial vehicles (UAVs), hence departing from conventional perimeter defences that may expose vulnerabilities. The Zero Trust Architecture (ZTA) paradigm requires a rigorous and continuous process of authenticating all network entities and communications. The accuracy of our methodology in detecting and identifying unmanned aerial vehicles (UAVs) is 84.59\%. This is achieved by utilizing Radio Frequency (RF) signals within a Deep Learning framework, a unique method. Precise identification is crucial in Zero Trust Architecture (ZTA), as it determines network access. In addition, the use of eXplainable Artificial Intelligence (XAI) tools such as SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) contributes to the improvement of the model's transparency and interpretability. Adherence to Zero Trust Architecture (ZTA) standards guarantees that the classifications of unmanned aerial vehicles (UAVs) are verifiable and comprehensible, enhancing security within the UAV field., Comment: 6 pages, 5 figures

Over the last few years, primordial black holes (PBHs) have emerged as a strong candidate for cold dark matter. A significant number of PBHs are produced when the strength of the primordial scalar power spectrum is enhanced on small scales (compared to the COBE normalized values on large scales). Such primordial spectra also inevitably lead to strong amplification of the scalar-induced, secondary gravitational waves (GWs) at higher frequencies. The recent detection of the stochastic gravitational wave background (SGWB) by the pulsar timing arrays (PTAs) has opened up the possibility of directly probing the very early universe. Different studies have shown that, when PBHs are assumed to have been formed during the epoch of radiation domination, the mechanism for the amplification of the scalar-induced GWs that is required to explain the PTA data can overproduce the PBHs over some ranges of masses. In this work, we assume a specific functional form for the primordial scalar power spectrum and examine the production of PBHs and the scalar-induced secondary GWs during the phase of reheating, which precedes the standard epoch of radiation domination. Specifically, we account for the uncertainties in the conditions for the formation of PBHs and ensure that the extent of PBHs produced remains within the observational bounds. We find that the scalar-induced SGWB generated during a phase of reheating with a steeper equation of state (than that of radiation) fit the NANOGrav 15-year data with a stronger Bayesian evidence than the astrophysical scenario involving GWs produced by merging supermassive binary black holes., Comment: 40 pages, 6 figures, 3 tables, published in JCAP

We present a novel method for imitation learning for control requirements expressed using Signal Temporal Logic (STL). More concretely we focus on the problem of training a neural network to imitate a complex controller. The learning process is guided by efficient data aggregation based on counter-examples and a coverage measure. Moreover, we introduce a method to evaluate the performance of the learned controller via parameterization and parameter estimation of the STL requirements. We demonstrate our approach with a flying robot case study.

We study offline multitask representation learning in reinforcement learning (RL), where a learner is provided with an offline dataset from different tasks that share a common representation and is asked to learn the shared representation. We theoretically investigate offline multitask low-rank RL, and propose a new algorithm called MORL for offline multitask representation learning. Furthermore, we examine downstream RL in reward-free, offline and online scenarios, where a new task is introduced to the agent that shares the same representation as the upstream offline tasks. Our theoretical results demonstrate the benefits of using the learned representation from the upstream offline task instead of directly learning the representation of the low-rank model., Comment: Accepted to 38th Conference on Neural Information Processing Systems (NeurIPS 2024)

We propose a deep learning based novel prediction framework for enhanced bandwidth reduction in motion transfer enabled video applications such as video conferencing, virtual reality gaming and privacy preservation for patient health monitoring. To model complex motion, we use the First Order Motion Model (FOMM) that represents dynamic objects using learned keypoints along with their local affine transformations. Keypoints are extracted by a self-supervised keypoint detector and organized in a time series corresponding to the video frames. Prediction of keypoints, to enable transmission using lower frames per second on the source device, is performed using a Variational Recurrent Neural Network (VRNN). The predicted keypoints are then synthesized to video frames using an optical flow estimator and a generator network. This efficacy of leveraging keypoint based representations in conjunction with VRNN based prediction for both video animation and reconstruction is demonstrated on three diverse datasets. For real-time applications, our results show the effectiveness of our proposed architecture by enabling up to 2x additional bandwidth reduction over existing keypoint based video motion transfer frameworks without significantly compromising video quality.

We explore the possibility of producing the observed matter-antimatter asymmetry of the Universe uniquely from the evaporation of primordial black holes (PBH) that are formed in an inflaton-dominated background. Considering the inflaton $(\phi)$ to oscillate in a monomial potential $V(\phi)\propto\phi^n$, we show, it is possible to obtain the desired baryon asymmetry via vanilla leptogenesis from evaporating PBHs of initial mass $\lesssim 10$ g. We find that the allowed parameter space is heavily dependent on the shape of the inflaton potential during reheating (determined by the exponent of the potential $n$), the energy density of PBHs (determined by $\beta$), and the nature of the coupling between the inflaton and the Standard Model (SM). To complete the minimal gravitational framework, we also include in our analysis the gravitational leptogenesis set-up through inflaton scattering via exchange of graviton, which opens up an even larger window for PBH mass, depending on the background equation of state. We finally illustrate that such gravitational leptogenesis scenarios can be tested with upcoming gravitational wave (GW) detectors, courtesy of the blue-tilted primordial GW with inflationary origin, thus paving a way to probe a PBH-induced reheating together with leptogenesis., Comment: 23 pages, 11 figures, 2 tables, accepted in PRD

Precisely measuring pressure in microfluidic flows is essential for flow control, fluid characterization, and monitoring, but faces specific challenges such as \RE{achieving} sufficient resolution, non-invasiveness, or ease of use. Here, we demonstrate a fully integrated multiplexed optofluidic pressure sensor, entirely decoupled from the flow path, that enables local pressure measurements along any microfluidic channel without altering its flow geometry. The sensor itself relies on the compression of a soft mechano-actuated hydrogel, changing color in response to a pressure change. The hydrogel is separated from the fluid circulating in the channel by a thin membrane, allowing for the unrestricted use of different types of fluids. Imaging the gel through the transparent PDMS with a color camera provides a direct, easy, and contact-free determination of the fluid pressure at the sensing location for pressures as small as \SI{20}{\milli\bar} with a resolution of around \SI{10}{\milli\bar}. The sensitivity and accessible pressure range can be tuned via the mechanical properties \RE{of the sensing unit}. The photonic gel can also be used to acquire 2D pressure or deformation maps, taking advantage of the fast response time and fine spatial resolution.

With a large number of populations, many problems are rising rapidly in Dhaka, the capital city of Bangladesh. Water-logging is one of the major issues among them. Heavy rainfall, lack of awareness and poor maintenance causes bad sewerage system in the city. As a result, water is overflowed on the roads and sometimes it gets mixed with the drinking water. To overcome this problem, this paper realizes the potential of using Internet of Things to combat water-logging in drainage pipes which are used to move wastes as well as rainwater away from the city. The proposed system will continuously monitor real time water level, water flow and gas level inside the drainage pipe. Moreover, all the monitoring data will be stored in the central database for graphical representation and further analysis. In addition to that if any emergency arises in the drainage system, an alert will be sent directly to the nearest maintenance office., Comment: Global Conference on Technology and Information Management

The COVID-19 pandemic caused by the coronavirus had a significant effect on social, economic, and health systems globally. The virus emerged in Wuhan, China, and spread worldwide resulting in severe disease, death, and social interference. Countries implemented lockdowns in various regions to limit the spread of the virus. Some of them were successful and some failed. Here, several factors played a vital role in their success. But mostly these factors and their correlations remained unidentified. In this paper, we unlocked those factors that contributed to the success of lockdown during the COVID-19 pandemic and explored the correlations among them. Moreover, this paper proposes several strategies to control any pandemic situation in the future. Here, it explores the relationships among variables, such as population density, number of infected, death, recovered patients, and the success or failure of the lockdown in different regions of the world. The findings suggest a strong correlation among these factors and indicate that the spread of similar kinds of viruses can be reduced in the future by implementing several safety measures.

Artificial intelligence has been around for a while, but suddenly it has received more attention than ever before. Thanks to innovations from companies like Google, Microsoft, Meta, and other major brands in technology. OpenAI, though, has triggered the button with its ground-breaking invention ChatGPT. ChatGPT is a Large Language Model (LLM) based on Transformer architecture that has the ability to generate human-like responses in a conversational context. It uses deep learning algorithms to generate natural language responses to input text. Its large number of parameters, contextual generation, and open-domain training make it a versatile and effective tool for a wide range of applications, from chatbots to customer service to language translation. It has the potential to revolutionize various industries and transform the way we interact with technology. However, the use of ChatGPT has also raised several concerns, including ethical, social, and employment challenges, which must be carefully considered to ensure the responsible use of this technology. The article provides an overview of ChatGPT, delving into its architecture and training process. It highlights the potential impacts of ChatGPT on the society. In this paper, we suggest some approaches involving technology, regulation, education, and ethics in an effort to maximize ChatGPT's benefits while minimizing its negative impacts. This study is expected to contribute to a greater understanding of ChatGPT and aid in predicting the potential changes it may bring about., Comment: 13 Pages

Emerging mobile virtual reality (VR) systems will require to continuously perform complex computer vision tasks on ultra-high-resolution video frames through the execution of deep neural networks (DNNs)-based algorithms. Since state-of-the-art DNNs require computational power that is excessive for mobile devices, techniques based on wireless edge computing (WEC) have been recently proposed. However, existing WEC methods require the transmission and processing of a high amount of video data which may ultimately saturate the wireless link. In this paper, we propose a novel Sensing-Assisted Wireless Edge Computing (SAWEC) paradigm to address this issue. SAWEC leverages knowledge about the physical environment to reduce the end-to-end latency and overall computational burden by transmitting to the edge server only the relevant data for the delivery of the service. Our intuition is that the transmission of the portion of the video frames where there are no changes with respect to previous frames can be avoided. Specifically, we leverage wireless sensing techniques to estimate the location of objects in the environment and obtain insights about the environment dynamics. Hence, only the part of the frames where any environmental change is detected is transmitted and processed. We evaluated SAWEC by using a 10K 360$^{\circ}$ with a Wi-Fi 6 sensing system operating at 160 MHz and performing localization and tracking. We considered instance segmentation and object detection as benchmarking tasks for performance evaluation. We carried out experiments in an anechoic chamber and an entrance hall with two human subjects in six different setups. Experimental results show that SAWEC reduces both the channel occupation and end-to-end latency by more than 90% while improving the instance segmentation and object detection performance with respect to state-of-the-art WEC approaches., Comment: Submitted to ACM for possible publication

We present an information-theoretic lower bound for the problem of parameter estimation with time-uniform coverage guarantees. Via a new a reduction to sequential testing, we obtain stronger lower bounds that capture the hardness of the time-uniform setting. In the case of location model estimation, logistic regression, and exponential family models, our $\Omega(\sqrt{n^{-1}\log \log n})$ lower bound is sharp to within constant factors in typical settings., Comment: 16 pages