Transfer Learning for Human & AI

The Transient objective is not to visit any state infinitely often . While this is not possible in any finite Markov Decision Process, it can be besatisfied in countably infinite ones, e.g., if the transition graph is acyclic . Optimal strategies for Transient need not exist, even if the MDP is infinitely branching .…

Affine phase retrieval arises in holography, dataseparation and phaseless sampling . Separable multi-block convex optimization problem appears in manymathematical and engineering fields . In the second part of this paper, we introduce a compressive affine phaseretrieval via lifting approach . We propose an algorithm torecover sparse real signals from their (noisy) affine quadratic measurements .…

In the classical dynamic BST model, the cost of both link traversal andbasic reconfiguration (rotation) is $O(1) In the matching model, two optical switches represent two matchings in an abstract way . In this work, we propose a simple dynamic BST algorithm that is based on dynamic Shannon-Fano-Elias coding .…

In this paper, we design and analyze new methods based on Additive AverageSchwarz–AAS Method . The new methods are designed forelliptic problems with highly heterogeneous coefficients . The methods are ofnon-overlapping type and subdomain interactions are obtained via the coarsespace .…

We consider the challenging problem of tracking multiple objects using adistributed network of sensors . The proposed algorithm is fast and requires significantly less processing time than fusion methods using multi-object filtering methods . It achieves better tracking accuracy by considering trackingerrors measured by the Optimal Sub-Pattern Assignment (OSPA) metric overseveral scans rather than a single scan.…

Researchers have used social media data to estimate various macroeconomic indicators about public behaviors . But the strong correlations disappeared when those models were tested with newer data according to a recent survey . By using decadal data (2008-2019)from Reddit, we show that both monthly and daily estimates of CCI can, indeed, be reliably estimated at least several months in advance, and that our modelestimates are far superior to those generated by the existing methods .…

This paper proposes a novel, rigorous and simple Fourier-transformationapproach to study resonances in a perfectly conducting slab with finite number of subwavelength slits of width $h\ll 1$. Since regions outside the slits are variable separated, by Fourier transforming the governing equation, we could express field in the outer regions in terms of field derivatives on theaperture .…

The method consists in verifying the components through successive levels, related to agreed-upon facets . It can be used in different domains where we can verifycomplex and heterogeneous systems based on their models, in order to avoid theloss of time and the over-cost of errors detection in an advanced phase .…

Network coding (NC) enables a linearprogramming (LP) formulation for a multi-source multicast with intra-sessionnetwork coding (MISNC) problem . However, it is still hard to solve using conventional methods due to the enormous scale of variables or constraints . In this paper, we try to solve this problem in terms of throughput maximizationfrom an algorithmic perspective .…

Novel well-balanced positivity preservingdiscretization techniques are proposed for the water surface elevation and the concentration of the pollutant . For the hydrodynamic system, the proposedscheme preserves the steady state of a lake at rest and the positivity of the water depth .…

In this paper, we consider the fundamental problem of approximation offunctions on a low-dimensional manifold embedded in a high-dimensional space . We propose a new approximation method that leverages theadvantages of the Manifold Locally Optimal Projection (MLOP) method and the strengths of the method ofRadial Basis Functions (RBF) The method is parametrization free, requires noknownowledge regarding the manifold’s intrinsic dimension, can handle noise andoutliers in both the function values and in the location of the data, and is applied directly in the high dimensions .…

Emerging application scenarios, such as cyber-physical systems (CPSs), theInternet of Things (IoT), and edge computing, call for coordination approaches . Field-based coordination is one such approach, promoting the idea of programming system coordination declaratively from a global perspective . We propose an alternative approach wherescheduling is programmed in a natural way (along with usual field-basedcoordination) in terms of causality fields .…

Determining the value of basketball players through analyzing the players’ behavior is important for the managers of modern basketball teams . Existing models based on dynamicnetwork theory offer major improvements to the results of such evaluations, but said models remain imprecise because they focus merely on evaluating the values of individual players rather than considering them within their current teams .…

The amount of sun cast on roads and parking lots determines the charging opportunities for solar vehicles and impacts the efficiency of conventional vehicles . We introduce a new approach to addressthese factors using pixel substitution and a light penetration factor .…

Graphtheory is a widely used framework for modeling complex engineered systems toevaluate their resilience to attacks . Resilience quantifies theability of the system to absorb and recover from extreme conditions . The proposed approach isaccurate in approximating the node resilience scores and offers a significant advantage over conventional approaches .…

In this paper, we adopt the fluid limits to analyze Age of Information (AoI) in a wireless multiaccess network with many users . We consider the case wherein users have heterogeneous i.i.d. channel conditions and the statuses aregenerate-at-will . We also use it to analyze the average non-linear AoI functions (with power and logarithm forms) in wireless networks .…

This paper proposes a robust deep learning framework used for classifyinganomaly of respiratory cycles . Front-end feature extraction step aims to transform the respiratory inputsound into a two-dimensional spectrogram . Next, an ensemble of C- DNN and Autoencodernetworks is then applied to classify into four categories of respiratoryanomaly cycles .…

Graph theory is widely used framework for modeling interdependent systems and to evaluatetheir resilience to disruptions . We propose a Bayesiangraph neural network-based framework for quickly identifying critical nodes in a large graph . Instead of using the observed graph for training the model, a MAP estimate of thegraph is computed based on the observed topology, and node target labels .…

We develop an analytic model of the mammalian cochlea . We use a mixedphysical-phenomenological approach . Spatial variation is incorporated through asingle independent variable that combines space and frequency . The model also predicts impedances that are qualitatively consistent with current literature .…

Traditional deep neural networks (NNs) have significantly contributed to the state-of-the-art performance in the task of classification under various domains . Unlike Bayesian NN that indirectly infer uncertainty through weightuncertainties, evidential NNs have been recently proposed to explicitlymodel the uncertainty of class probabilities and use them for classification tasks .…

This paper deals with the modeling and estimation of the sociologicalphenomena called echo chambers and segregation in social networks . We present a novel community-based graph model that representssthe emergence of segregated echo chambers as a Markov bridge process . A Markovbridge is a one-dimensional Markov random field that facilitates modeling theformation and disassociation of communities at deterministic times .…

The heterogeneity of tools that support temporal logic formulae poses several challenges in terms of interoperability . This document proposes standardgrammars for Linear Temporal Logic (LTL) and Linear Dynamic Logic .…

We study the idea of variance reduction applied to adaptive stochastic mirrordescent algorithms in nonsmooth nonconvex finite-sum optimization problems . Wepropose a simple yet generalized adaptive mirror descent algorithm withvariance reduction named SVRAMD . We prove that variance reduction reduces the gradientcomplexity of most adaptive mirrors descent algorithms and boost theirconvergence.…

In this paper, based on the idea of self-adjusting steepness basedschemes, a two-dimensional calculation method of steepness parameter is proposed . With the application of such limiter to the over-intersectionbased remapping framework, a low dissipation remapping method has been proposed .…

The maximum possible throughput (or the rate of job completion) of amulti-server system is typically the sum of the service rates of individual servers . Launching multiple replicas of a job and canceling them as soon as one copy finishes can boost the throughput, especially when the service time distribution has high variability .…

Maximum run-length limited codes are constraint codes used in communication and data storage systems . Insertion/deletion correcting codes correct insertionor deletion errors caused in transmitted sequences and are used for combatingsynchronization errors . This paper investigates RLL-SIDC codes, which are efficient encodable and decodable codes .…

Cell-free massive MIMO provides ubiquitous connectivity for multiple users . The major cost includes fronthaul overheads and APhardware . Maximum ratio combination (MRC) achieves a low frontthaul loading andlow-cost AP, but the performance is bad . This letter proposes to implement aquasi-LMMSE (Q-LmmSE) processing using MRC fron thaul design .…

The quality of a test collection greatly depends on the number of participants and the quality of the submitted runs . Results suggest that test collections can be predicted prior to collecting re-evaluation judgments, and that quality of tests can be inferred prior to assessing relevance of a given document .…

Grant-free multiple-input multiple- input multiple-output (MIMO) usually employsnon-orthogonal pilots for joint user detection and channel estimation . But existing methods are too complex for massive grant-free access in massive MIMO . This letter proposes pilot detection reference signal (PDRS) to greatly reduce the complexity .…

The upcomingsixth-generation (6G) wireless network is required to support grant-free random access of a massive number of sporadic traffic devices . The paper proposes a novel deep learning framework for JADCE in 6G wireless networks . Prior-feature learning followedby an adaptive-tuning strategy is proposed, where an inner network composed of the Expectation-maximization (EM) and back-propagation is introduced to jointlytune the precision and learn the precision of the device statematrix .…

A brain-inspired search engine assistant named DeveloperBot based on knowledge graph is proposed . It aligns to the cognitive process of human and has the capacity to answer complex queries withexplainability . The results of the decision-making demonstrate that DeveloperBotcan estimate the answers and answer confidences with high accuracy .…

We propose an accurate algorithm for a novel sum-of-exponentials (SOE)approximation of kernel functions . We also develop a fast algorithm for convolutionquadrature based on the SOE . The SOE method is efficient and accurate, and works for general kernels with a controlable upperbound of positive exponents .…

The financial model proposed involves the liquidation process of a portfolio of $n$ assets through sell or (and) buy orders with volatility . In dimensions $n=3$ and for zero volatility, this problem standsas a mean field model for Ostwald ripening, and has been proposed and analyzed by Niethammer .…

Self-supervised representation learning of biological sequence embeddings can be useful for pretraining encoders . Viewingevolution as natural sequence augmentation and maximizing information acrossphylogenetic “noisy channels” is a biologically and theoretically desirable objective . We provide an illustrative example where contrastive learning using evolutionary augmentation can be used as arepresentation learning objective which maximizes the mutual informationbetween biological sequences and their conserved function, and finally outlinerationale for this approach .…

A workflow-net with Actors (WFA-net) has been developed to tackle the problem of information quality . WFA-Net allows for IQ requirements in their social and organizational context . Thispaper reports on the experience gained, findings and lessons learned whiledeveloping the WFA -net .…

Tr-centrality is a centrality measure which focuses on using thenode triangle structure and the node neighborhood information to define the strength of a node . It is greatly significant in evaluating nodes Influence ranking in complexnetworks . Toverify the validity of Tr-Centrality, we apply it to four real-world networks with different densities and shapes and it has proven to yield better results .…

The MCMF ADL-MVDR handles different numbers of microphone channels in one framework . Spatio-temporalcross correlations are also fully utilized in the proposed approach . The proposed system is evaluated using a Mandarin audio-visual corpora and iscompared with several state-of-the-art approaches .…

A wide variety of sensor technologies are nowadays used for traffic monitoring applications . Most of these technologies rely on wired infrastructure, the installation and maintenance costs limit the deployment of traffic monitoring systems . In this paper, we introduce a trafficmonitoring approach that exploits dedicated short-range communications (DSRC)signals sent in a vehicular network and machine learning techniques .…

Fuzzing has become the primetechnique to detect bugs and vulnerabilities in programs . To generate inputsthat cover new functionality, fuzzers require execution feedback from theprogram . If such feedback is not available, fuzzing can only rely on chance, which isineffective .…

Diseases spread over temporal networks of contacts between individuals . Structures of these temporal networks hold the keys to understanding epidemicpropagation . Early concept of the literature to aid in discussing thesestructures is concurrency — quantifying individuals’ tendency to formtime-overlapping “partnerships” An emerging body of literature is trying to connect methods to the concurrency literature .…

Quantum computing could have exponential speed-up in comparison to their classical counterparts . In this paper, we propose an improved version of aquantum edge detection algorithm . The application of quantum computing to the field of image processing has several promising applications, such as quantum image processing, has been developed in the past .…

In this article, we propose apatch reconstruction finite element space with only one degree of freedom perelement . It is applied to the discontinuousGalerkin methods with the upwind scheme for the steady-stateconvection-diffusion-reaction problems over polytopic meshes . The optimal errorestimates are provided in both diffusion-dominated and convection-dominated regimes .…

More than 50 countries have developed COVID contact-tracing apps to limit the spread of coronavirus . However, many experts and scientists cast doubt on effectiveness of those apps . For each app, a large number of reviews have been entered by end-users in app stores .…

The use of Terahertz frequency bands for data transmissions between thecore network and an access point can be promising for next generation wireless systems . We analyze the performance of a dual-hop relaying forTHz-RF wireless link for backhaul applications . Using analytical results of the direct link and computer simulations, we demonstrate that theTHzRF relaying is a viable technology for wireless backhaul, providing asignificant increase of almost $25 \%$ in the spectral efficiency, compared to the direct transmissions .…

Domain Name System (DNS) is a crucial component of current IP-based networks . Due to its lack of data integrity and origin authentication processes, it is vulnerable to a variety of attacks . Detecting this attack is particularly important as it can be a threat to corporate secrets and can beused to steal information or commit fraud .…

The directional radius of a cell is defined as the distance from the nucleus to the cell boundary at an angle relative to the direction of a uniformly randomlocation in the cell . We study the distribution of the radii in two types of cell types in the Poisson Voronoi tessellations .…

This paper proposes a deep learning approach to the adaptive and sequential beamforming design problem for the initial access phase in a mmWave environment . The problem is equivalent to designing the sequence of sensing beamformers to learn the angleof arrival (AoA) of the dominant path .…

A high success rate of grant-free random access scheme is proposed to support massive access for machine-to-machine communications in massive multiple input and output systems . This scheme allows active user equipments (UEs) totransmit their modulated uplink messages along with super pilots consisting of multiple sub-pilots to a base station .…

Quantum networks use quantum mechanics properties of entanglement and teleportation to transfer data from one node to another . It is necessary to have an efficient mechanism to distribute entanglements among quantum networknodes . Most of research on quantum networks apply current state of network and do not consider using historical data .…

An LSTM-aided hybrid random access scheme (LSTMH-RA) is proposed to support diverse quality of service (QoS) requirements in 6G MTC heterogeneous networks . This scheme employs an attention-basedLSTM prediction model to predict the number of active URLLC devices, determinest the parameters of the multi-user detection algorithm dynamically, and then allows URllC devices to access the network via a two-step contention-free procedure, to meet latency and reliability access requirements .…