Abstract

A large portion of the computation in sparse neural networks comprises of multiplying a sparse matrix with a dense matrix, denoted SDMM in this paper. The SDMM operation with an unstructured sparsity pattern cannot be efficiently processed on modern architectures such as GPUs due to irregularity in compute and memory accesses. However, efficient parallel algorithms on a GPU can be designed for SDMM when the sparsity pattern is more structured. Thus, the run time performance of sparse neural networks on a GPU is dependent on the sparsity pattern present in the underlying matrix.In sparse neural networks obtained using pruning based approaches, the choice of sparsity pattern not only effects the run time, but also effects the accuracy of the task for which the neural network is trained for. Sparsity patterns which have a good run time performance on a GPU may not have a good accuracy and vice-versa. The real challenge then is to given a target architecture, identify a sparsity pattern, a storage format,and an algorithm for SDMM that leads to sparse neural networks which are efficient in both run time and accuracy.In this work, we propose a novel, structured, flexible, and generic sparsity pattern called the RMB (Regularized MultiBlock) sparsity pattern, and an efficient storage format (CRMB),and a fast GPU algorithm for processing RMBMM (SDMMwith the multiplicand having RMB sparsity pattern). Using theRMB sparsity pattern, we achieve better trade-offs between the accuracy, and the run time performance of sparse neural networks on a GPU when compared to commonly used sparsity patterns like unstructured, and block sparsity patterns.

Abstract

In this paper, we study scalable parallel algorithms for estimating the farness-centrality value of the nodes in a given undirected and connected graph. Our algorithms consider approaches that are more suitable for sparse graphs. To this end, we propose four optimization techniques based on removing redundant nodes, removing identical nodes, removing chain nodes, and making use of decomposition based on the biconnected components of the input graph.We test our techniques on a collection of real-world graphs for the time taken and the average error percentage. We further analyze the applicability of our techniques on various classes of real-world graphs. We suggest why certain techniques work better on certain classes of graphs

Abstract

Quality Models play a critical role in assuring quality and have evolved over 40+ years. They provide support for defining quality attributes, building and measuring the quality of the resulting product. Each quality model adopts a critical view on quality in terms of a set of model elements and relationships between them. This study aims to provide an overview of the state-of-the-art research on quality models with a focus on encompassing model elements and their support to architecting quality. The study was conducted using systematic mapping as the research methodology. A total of 238 primary papers were classified based on the type of research, standards usage, and publication trends. We identified that 17% (40) of papers belong to quality models. These 40 models were analyzed for the underlying meta-model elements and their support for a quality architecture using   rence architecture framework. The architecture phase mapping analysis shows that quality planning phase is 100% supported, quality assessment is 75% supported, quality documentation is included in 40% models and quality realization aspect is barely considered in 13% models. Quality realization happens through software processes and patterns, and it is necessary to evolve quality models and software process architectures that correlate quality definitions and quality realization mechanisms. Future research is expected in this direction

Abstract

We focus on the problem of symmetric key distribution for securing shared resources among large groups of users in distributed applications like cloud storage, shared databases, and collaborative editing, among others. In such applications, resources such as data, are sensitive in nature and it is necessary that only authorized users are allowed access without the presence of on-line monitoring system. The de-facto approach is to encrypt a shared resource and deploy a key distribution mechanism, which enables only authorized users to generate the respective decryption key for the resource. The key distribution approach has two major challenges: first, the applications are dynamic i.e., users might join and leave arbitrarily, and second, for a large number of users, it is required that the cryptographic technique be scalable and efficient. In this work, we describe an approach that overcomes these challenges by using two key techniques: first, flattening the access structure and applying efficient symmetric key distribution techniques. By flattening the access structure, we reduce the problem to that of key distribution of a resource among all the users sharing that resource. We consider this smaller flattened access structure and devise a unified key distribution technique that is sufficient for key distribution across all such structures. Our key distribution techniques have an important feature of a public secret and a private secret, which allows the group controller to publish updates to the keying material using the public secret and therefore, does not necessitate the users to be in constant communication with the group controller. Using this model we describe two efficient key distribution techniques that scale logarithmically with the group size and also handle group additions and removals. Furthermore, a user can be off-line for any amount of time and need not be aware of the dynamics of the system, which is important as it overcomes the problems posed by lossy channels. We have performed an experimental evaluation of our scheme against a popular existing scheme and show that they perform better for this scheme with the same security guarantees. As our approaches are easy to implement they are especially suitable for practical applications where security is viewed as an overhead rather than as a necessity.

Abstract

Building accurate lexicon free handwritten text recognizers for Indic languages is a challenging task, mostly due to the inherent complexities in Indic scripts in addition to the cursive nature of handwriting. In this work, we demonstrate an end-to-end trainable CNN-RNN hybrid architecture which takes inspirations from recent advances of using residual blocks for training convolutional layers, along with the inclusion of spatial transformer layer to learn a model invariant to geometric distortions present in handwriting. In this work we focus building state of the art handwritten word recognizers for two popular Indic scripts – Devanagari and Bangla. To address the need of large scale training data for such low resources languages, we utilize synthetically rendered data for pre-training the network and later fine tune it on the real data. We outperform the previous lexicon based, state of the art methods on the test set …

Abstract

Weighted Frequent Itemset (WFI) mining is animportant model in data mining. It aims to discover all itemsetswhose weighted sum in a transactional database is no less than theuser-specified threshold value. Most previous works focused onfinding WFIs in a transactional database and did not recognizethe spatiotemporal characteristics of an item within the data.This paper proposes a more flexible model of Spatial WeightedFrequent Itemset (SWFI) that may exist in a spatiotemporaldatabase. The recommended patterns may be found very useful inmany real-world applications. For instance, an SWFI generatedfrom an air pollution database indicates a geographical regionwhere people have been exposed to high levels of an air pollutant,sayPM2.5. The generated SWFIs do not satisfy the anti-monotonic property. Two new measures have been presentedto effectively reduce the search space and the computationalcost of finding the desired patterns. A pattern-growth algorithm,called Spatial Weighted Frequent Pattern-growth, has also beenpresented to find all SWFIs in a spatiotemporal database.Experimental results demonstrate that the proposed algorithmis efficient. We also describe a case study in which our modelhas been used to find useful information in air pollution database.

Abstract

In this paper, we present an algorithmic frame-work which computes the collision-free velocities for the robotin a human shared dynamic and uncertain environment. Weextend the concept of Inverse Velocity Obstacle (IVO) to aprobabilistic variant to handle the state estimation and motionuncertainties that arise due to the other participants of the en-vironment. These uncertainties are modeled as non-parametricprobability distributions. In our PIVO: Probabilistic InverseVelocity Obstacle, we propose the collision-free navigation asan optimization problem by reformulating the velocity condi-tions of IVO as chance constraints that takes the uncertaintyinto account. The space of collision-free velocities that resultfrom the presented optimization scheme are associated to aconfidence measure as a specified probability. We demonstratethe efficacy of our PIVO through numerical simulations anddemonstrating its ability to generate safe trajectories underhighly uncertain environments

Abstract

n this article, we propose exact passive-aggressive(PA) online algorithms for ordinal regression. The proposedalgorithms can be used even when we have interval labelsinstead of actual labels for example. The proposed algorithmssolve a convex optimization problem at every trial. We find anexact solution to those optimization problems to determine theupdated parameters. We propose a support class algorithm (SCA)that finds the active constraints using the Karush–Kuhn–Tucker(KKT) conditions of the optimization problems. These active con-straints form a support set, which determines the set of thresholdsthat need to be updated. We derive update rules for PA, PA-I, and PA-II. We show that the proposed algorithms maintainthe ordering of the thresholds after every trial. We provide themistake bounds of the proposed algorithms in both ideal andgeneral settings. We also show experimentally that the proposedalgorithms successfully learn accurate classifiers using intervallabels as well as exact labels. The proposed algorithms also dowell compared to other approaches.

Abstract

The commercial sector accounts for 8.6% of total electricity consumption of India and increasing with 5% rate annually due to rapid urbanization. The building's major energy consuming end-uses are air-conditioning including heating, cooling, lighting, and equipment. The development and implementation of energy efficiency codes and measures will provide sustainable future. The development process requires analyses of current building construction and operation practices. The current building construction practices can be analyzed by developing reference buildings representative of national building stock. In this research, the development methodology of reference building, developed reference office buildings for India and its application have been presented. The reference buildings are developed for low-rise and high-rise office buildings with 8-h and 24-h operation based on data collection of 230 buildings constructed in last ten years. This methodology can be applied to other buildings types for development of reference buildings with the appropriate data. From the analyses of reference buildings, it has found that the primary energy conservation measures would be envelope thermal properties for 8-h operational buildings while internal loads (occupancy, lighting, and equipment) in 24-h operational building. The energy performance index (EPI) of reference office buildings are better than ECBC 2017. Besides, some of the minimum requirements have not met. The lighting and glass specifications of reference buildings are better than ECBC level due market and government policies while HVAC system specially chiller COP and its controls are needed to be improved.

Abstract

Forecasting energy consumption enables users to plan their resource utilization optimally. For this, it is essential to make reliable forecasting of consumption profile in real-time which is very challenging and still emerging. In this study, we focus on forecasting HVAC cooling energy consumption at sub-hourly forecasting horizon which enables us to analyze and control the demand in real-time. With the aid of energy meters and adaptation of BACnet technology in the target building, we collected real building cooling energy consumption data and HVAC systems usage at a higher resolution. In this methodology, weather data from Weather Underground of Gachibowli, Hyderabad location has been used. These recorded weather parameters are being used as explanatory variables. Along with the weather parameters, many other feature-engineered explanatory variables which are discussed in the paper has been used to capture the trend and variation of energy consumption. Required pre-processing and data analysis is performed on the raw data to obtain meaningful information for modeling. We have implemented one-step ahead static forecasting using Gradient Boosting Regression Trees (GBRT), where we forecast for single-time-step ahead in each iteration then update the training data at the end of each iteration with actual data available next time-step in order to assess how well the model forecasts one-step-ahead. Our results show that we could improve forecast accuracy by almost 45% by including feature engineered variables. We achieved a Mean Absolute Percentage Error (MAPE) of 14.7% with GBRT.