Search

The explosion of IoT sensors in industrial, consumer and remote sensing use cases has come with unprecedented demand for computing infrastructure to transmit and to analyze petabytes of data. Concurrently, the world is slowly shifting its focus towards more sustainable computing. For these reasons, there has been a recent effort to reduce the footprint of related computing infrastructure, especially by deep learning algorithms, for advanced insight generation. The `TinyML' community is actively proposing methods to save communication bandwidth and excessive cloud storage costs while reducing algorithm inference latency and promoting data privacy. Such proposed approaches should ideally process multiple types of data, including time series, audio, satellite images, and video, near the network edge as multiple data streams has been shown to improve the discriminative ability of learning algorithms, especially for generating fine grained results. Incidentally, there has been recent work on data driven conditional computation of subnetworks that has shown real progress in using a single model to share parameters among very different types of inputs such as images and text, reducing the computation requirement of multi-tower multimodal networks. Inspired by such line of work, we explore similar per patch conditional computation for the first time for mobile vision transformers (vision only case), that will eventually be used for single-tower multimodal edge models. We evaluate the model on Cornell Sap Sucker Woods 60, a fine grained bird species discrimination dataset. Our initial experiments uses $4X$ fewer parameters compared to MobileViTV2-1.0 with a $1$% accuracy drop on the iNaturalist '21 birds test data provided as part of the SSW60 dataset.

The use of social media applications, hate speech engagement, and public debates among teenagers, primarily by university and college students, is growing day by day. The feelings of tremendous stress, anxiety, and depression via social media among our youths have a direct impact on their daily lives and personal workspace apart from delayed sleep, social media addictions, and memory loss. The use of NO phone times and NO phone zones is now popular in workplaces and family cultures. The use of hate speech, negotiations, and toxic words can lead to verbal abuse and cybercrime. Growing concern of mobile device security, cyberbullying, ransomware attacks, and mental health issues are another serious impact of social media among university students. The future challenges including health issues of social media use and hate speech has a serious impact on livelihood, freedom, and diverse communities of university students. Our case study is related to social media use and hate speech related to public debates over university students. We have presented the analysis and impact of social media and hate speech with several conclusions, cybercrimes, and components. The use of questionnaires for collecting primary data over university students help in the analysis of case study. The conclusion of case study and future scope of the research is extremely important to counter negative impacts., Comment: 12 pages, 7 figures

This paper shows how a family of function spaces (coined as Assiamoua spaces) plays a fundamental role in the Fourier analysis of vector-valued functions compact groups. These spaces make it possible to transcribe the classic results of Fourier analysis in the framework of analysis of vector-valued functions and vector measures. The construction of Sobolev spaces of vector-valued functions on compact groups rests heavily on the members of the aforementioned family., Comment: In memory of VSK Assiamoua

Quantum Machine Learning (QML) is a red-hot field that brings novel discoveries and exciting opportunities to resolve, speed up, or refine the analysis of a wide range of computational problems. In the realm of biomedical research and personalized medicine, the significance of multi-omics integration lies in its ability to provide a thorough and holistic comprehension of complex biological systems. This technology links fundamental research to clinical practice. The insights gained from integrated omics data can be translated into clinical tools for diagnosis, prognosis, and treatment planning. The fusion of quantum computing and machine learning holds promise for unraveling complex patterns within multi-omics datasets, providing unprecedented insights into the molecular landscape of lung cancer. Due to the heterogeneity, complexity, and high dimensionality of multi-omic cancer data, characterized by the vast number of features (such as gene expression, micro-RNA, and DNA methylation) relative to the limited number of lung cancer patient samples, our prime motivation for this paper is the integration of multi-omic data, unique feature selection, and diagnostic classification of lung subtypes: lung squamous cell carcinoma (LUSC-I) and lung adenocarcinoma (LUAD-II) using quantum machine learning. We developed a method for finding the best differentiating features between LUAD and LUSC datasets, which has the potential for biomarker discovery., Comment: 27 pages, 17 figures

Symmetry is a powerful tool for finding analytical solutions to differential equations, both partial and ordinary, via the similarity variables or via the invariance of the equation under group transformations. It is the largest group of transformations that leaves the differential equation invariant. It is now known that this differential equation method plays the same role when it comes to the study of difference equations. Difference equations can be used to model various phenomena where the changes occur in discrete manner. The use of symmetries on recurrence equations, usually, leads to reductions of order and hence ease the process of finding their solutions. One of the aims of this work is to employ symmetries to generalize some results in the literature. We present new generalized formula solutions of a class of difference equations and we investigate the periodicity and behavior of theses solutions., Comment: 16

Burgers' equation with fixed Dirichlet boundary conditions is considered on generic bounded intervals. By using the Hopf-Cole transformation and the exact operational solution recently established for linear reaction-diffusion equations with constant coefficients, an exact solution in the time domain is implicitly derived by means of inverse Laplace transforms. Analytic inverses, whenever they exist, can be obtained in closed form using Mellin transforms. However, highly efficient algorithms are available, and numerical inverses in the time domain are always possible, regardless of the complexity of the Laplace domain expressions. Two illustration tests show that the results coincide well with those of classical exact solutions. Compared to the solutions obtained with series expressions or by numerical methods, closed-form expressions, even in the Laplace domain, represent an innovative alternative and new perspectives can be envisaged. The exact solution via the inverse Laplace transform is shown to be more computationally efficient and thus provides a reference point for numerical and semi-analytical methods., Comment: 15

Image environments and noisy labels hinder deep learning-based inference models in structural damage detection. Post-detection, there is the challenge of reliance on manual assessments of detected damages. As a result, Guided-DetNet, characterized by Generative Attention Module (GAM), Hierarchical Elimination Algorithm (HEA), and Volumetric Contour Visual Assessment (VCVA), is proposed to mitigate complex image environments, noisy labeling, and post-detection manual assessment of structural damages. GAM leverages cross-horizontal and cross-vertical patch merging and cross foreground-background feature fusion to generate varied features to mitigate complex image environments. HEA addresses noisy labeling using hierarchical relationships among classes to refine instances given an image by eliminating unlikely class categories. VCVA assesses the severity of detected damages via volumetric representation and quantification leveraging the Dirac delta distribution. A comprehensive quantitative study, two robustness tests, and an application scenario based on the PEER Hub Image-Net dataset substantiate Guided-DetNet's promising performances. Guided-DetNet outperformed the best-compared models in a triple classification task by a difference of not less than 3% and not less than 2% in a dual detection task under varying metrics.

In this paper, we introduce a family of Fourier multipliers using the spherical Fourier transform on Gelfand pairs. We refer to them as spherical Fourier multipliers. We study certain sufficient conditions under which they are bounded. Then, under the hypothesis of compactness of the underlying group and under certain summability conditions, we obtain the belonging of the spherical Fourier multipliers to some Schatten-von Neumann classes., Comment: 13 pages

Network slicing enables industrial Internet of Things (IIoT) networks with multiservice and differentiated resource requirements to meet increasing demands through efficient use and management of network resources. Typically, the network slice orchestrator relies on demand forecasts for each slice to make informed decisions and maximize resource utilization. The new generation of Industry 4.0 has introduced digital twins to map physical systems to digital models for accurate decision-making. In our approach, we first use graph-attention networks to build a digital twin environment for network slices, enabling real-time traffic analysis, monitoring, and demand forecasting. Based on these predictions, we formulate the resource allocation problem as a federated multi-agent reinforcement learning problem and employ a deep deterministic policy gradient to determine the resource allocation policy while preserving the privacy of the slices. Our results demonstrate that the proposed approaches can improve the accuracy of demand prediction for network slices and reduce the communication overhead of dynamic network slicing., Comment: 8 pages, 7 figures, conference

This paper deals with a class of Sobolev spaces of vector-valued functions on a compact group. Some Sobolev embedding theorems are proved., Comment: 7 pages

A kind of generalized Gelfand pair is introduced via a Banach algebra consisting of bi-invariant functions in a weighted Lebesgue space. The related spherical functions and the Fourier transformation are constructed. The multipliers of the underlying algebra are characterized by this Fourier transformation., Comment: 15 pages

The African continent lacks enough qualified teachers which hampers the provision of adequate learning support. An AI could potentially augment the efforts of the limited number of teachers, leading to better learning outcomes. Towards that end, this work describes and evaluates the first key output for the NSMQ AI Grand Challenge, which proposes a robust, real-world benchmark for such an AI: "Build an AI to compete live in Ghana's National Science and Maths Quiz (NSMQ) competition and win - performing better than the best contestants in all rounds and stages of the competition". The NSMQ is an annual live science and mathematics competition for senior secondary school students in Ghana in which 3 teams of 2 students compete by answering questions across biology, chemistry, physics, and math in 5 rounds over 5 progressive stages until a winning team is crowned for that year. In this work, we built Brilla AI, an AI contestant that we deployed to unofficially compete remotely and live in the Riddles round of the 2023 NSMQ Grand Finale, the first of its kind in the 30-year history of the competition. Brilla AI is currently available as a web app that livestreams the Riddles round of the contest, and runs 4 machine learning systems: (1) speech to text (2) question extraction (3) question answering and (4) text to speech that work together in real-time to quickly and accurately provide an answer, and then say it with a Ghanaian accent. In its debut, our AI answered one of the 4 riddles ahead of the 3 human contesting teams, unofficially placing second (tied). Improvements and extensions of this AI could potentially be deployed to offer science tutoring to students and eventually enable millions across Africa to have one-on-one learning interactions, democratizing science education., Comment: 14 pages. Accepted for the WideAIED track at the 25th International Conference on AI in Education (AIED 2024)

To address the learning crisis and other challenges facing children and youth (those 15-24 years of age) it is critical for governments, multilateral and bilateral funders, other donors, and implementers to have a clear understanding of program costs and cost breakdowns to inform financing decisions. For nearly a decade, the Center for Universal Education (CUE) at the Brookings Institution has led the Childhood Cost Data initiative, a research project focused on the collection, analysis, and use of cost data to increase the volume and quality of funding for programs targeting children and youth. The initiative's end goal is to improve learning and life outcomes for the youngest generation through these advancements. As part of this initiative, Brookings developed the Childhood Cost Calculator (C3), intended to facilitate cost analysis. This study introduces C3, the tool's various dimensions and functionalities, and presents learnings from three case studies from piloting the calculator in the education or early childhood development (ECD) sectors in Cambodia, Ghana, and Honduras. It concludes with suggestions for a way forward that supports funders, implementers, and researchers to better use cost data to make strides in tackling the multiple crises facing children and youth today.

In this paper, we study the higher-order Beverton-Holt equation. We derive nontrivial symmetries, and thereafter, solutions are obtained. For constant rate and carrying capacity, we study the periodic nature of the solution and analyze the stability of the equilibrium points have been analyzed., Comment: 11 pages, 6 figures

We present a condition under which the thermal quasi-geostrophic (TQG) model possesses a solution that is holomorphic in time with values in the Gevrey space of complex analytic functions. This can be seen as the complex extension of the work by Levermore and Oliver (1997) for the generalized Euler equation but applied to the TQG model., Comment: 11 pages

The system under study is a solute-solvent-structure (SSS) interaction problem for the interaction of a dilute three-dimensional Oldroyd-B polymeric fluid with a two-dimensional viscoelastic shell. We show that a unique global strong solution to this system exists under the condition that the classical Ladyzhenskaya--Prodi--Serrin criterion holds for the velocity field and that the shell displacement is essentially bounded in time with values in the space of continuously differentiable functions. No requirement is needed for the polymer number density and the extra stress tensor for the solute component., Comment: 20 pages