total 46took 0.07s

Multiple-Population Moment Estimation: Exploiting Inter-Population Correlation for Efficient Moment Estimation in Analog/Mixed-Signal ValidationMar 31 2014Moment estimation is an important problem during circuit validation, in both pre-Silicon and post-Silicon stages. From the estimated moments, the probability of failure and parametric yield can be estimated at each circuit configuration and corner, and ... More

Towards Gradient Free and Projection Free Stochastic OptimizationOct 08 2018Feb 18 2019This paper focuses on the problem of \emph{constrained} \emph{stochastic} optimization. A zeroth order Frank-Wolfe algorithm is proposed, which in addition to the projection-free nature of the vanilla Frank-Wolfe algorithm makes it gradient free. Under ... More

Multi-step Retriever-Reader Interaction for Scalable Open-domain Question AnsweringMay 14 2019This paper introduces a new framework for open-domain question answering in which the retriever and the reader iteratively interact with each other. The framework is agnostic to the architecture of the machine reading model, only requiring access to the ... More

Question Answering on Knowledge Bases and Text using Universal Schema and Memory NetworksApr 27 2017Existing question answering methods infer answers either from a knowledge base or from raw text. While knowledge base (KB) methods are good at answering compositional questions, their performance is often affected by the incompleteness of the KB. Au contraire, ... More

Deep SetsMar 10 2017Apr 14 2018We study the problem of designing models for machine learning tasks defined on \emph{sets}. In contrast to traditional approach of operating on fixed dimensional vectors, we consider objective functions defined on sets that are invariant to permutations. ... More

Spectral Methods for Nonparametric ModelsMar 31 2017Nonparametric models are versatile, albeit computationally expensive, tool for modeling mixture models. In this paper, we introduce spectral methods for the two most popular nonparametric models: the Indian Buffet Process (IBP) and the Hierarchical Dirichlet ... More

Point Cloud GANOct 13 2018Generative Adversarial Networks (GAN) can achieve promising performance on learning complex data distributions on different types of data. In this paper, we first show a straightforward extension of existing GAN algorithm is not applicable to point clouds, ... More

The Myths of Our Time: Fake NewsAug 05 2019While the purpose of most fake news is misinformation and political propaganda, our team sees it as a new type of myth that is created by people in the age of internet identities and artificial intelligence. Seeking insights on the fear and desire hidden ... More

Randomized Exploration in Generalized Linear BanditsJun 21 2019We study two randomized algorithms for generalized linear bandits, GLM-TSL and GLM-FPL. GLM-TSL samples a generalized linear model (GLM) from the Laplace approximation to the posterior distribution. GLM-FPL, a new algorithm proposed in this work, fits ... More

Go for a Walk and Arrive at the Answer: Reasoning Over Paths in Knowledge Bases using Reinforcement LearningNov 15 2017Dec 30 2018Knowledge bases (KB), both automatically and manually constructed, are often incomplete --- many valid facts can be inferred from the KB by synthesizing existing information. A popular approach to KB completion is to infer new relations by combinatory ... More

Developing Creative AI to Generate Sculptural ObjectsAug 20 2019We explore the intersection of human and machine creativity by generating sculptural objects through machine learning. This research raises questions about both the technical details of automatic art generation and the interaction between AI and people, ... More

Investigating the Working of Text ClassifiersJan 19 2018Aug 05 2018Text classification is one of the most widely studied tasks in natural language processing. Motivated by the principle of compositionality, large multilayer neural network models have been employed for this task in an attempt to effectively utilize the ... More

Federated Optimization for Heterogeneous NetworksDec 14 2018Jan 25 2019Federated learning involves training machine learning models in massively distributed networks. While Federated Averaging~(\fedavg) is the leading optimization method for training non-convex models in this setting, its behavior is not well understood ... More

Transformation Autoregressive NetworksJan 30 2018Oct 23 2018The fundamental task of general density estimation $p(x)$ has been of keen interest to machine learning. In this work, we attempt to systematically characterize methods for density estimation. Broadly speaking, most of the existing methods can be categorized ... More

Nonparametric Density Estimation under Adversarial LossesMay 22 2018Oct 28 2018We study minimax convergence rates of nonparametric density estimation under a large class of loss functions called "adversarial losses", which, besides classical $\mathcal{L}^p$ losses, includes maximum mean discrepancy (MMD), Wasserstein distance, and ... More

Compressed Video Action RecognitionDec 02 2017Mar 29 2018Training robust deep video representations has proven to be much more challenging than learning deep image representations. This is in part due to the enormous size of raw video streams and the high temporal redundancy; the true and interesting signal ... More

Open Domain Question Answering Using Early Fusion of Knowledge Bases and TextSep 04 2018Open Domain Question Answering (QA) is evolving from complex pipelined systems to end-to-end deep neural networks. Specialized neural models have been developed for extracting answers from either text alone or Knowledge Bases (KBs) alone. In this paper ... More

State Space LSTM Models with Particle MCMC InferenceNov 30 2017Long Short-Term Memory (LSTM) is one of the most powerful sequence models. Despite the strong performance, however, it lacks the nice interpretability as in state space models. In this paper, we present a way to combine the best of both worlds by introducing ... More

A Generic Approach for Escaping Saddle pointsSep 05 2017A central challenge to using first-order methods for optimizing nonconvex problems is the presence of saddle points. First-order methods often get stuck at saddle points, greatly deteriorating their performance. Typically, to escape from saddles one has ... More

Hallucinating Point Cloud into 3D Sculptural ObjectNov 13 2018Nov 29 2018Our team of artists and machine learning researchers designed a creative algorithm that can generate authentic sculptural artworks. These artworks do not mimic any given forms and cannot be easily categorized into the dataset categories. Our approach ... More

Federated Optimization for Heterogeneous NetworksDec 14 2018Jul 11 2019Federated learning involves training machine learning models in massively distributed networks. While Federated Averaging (FedAvg) is the leading optimization method for training non-convex models in this setting, its behavior is not well understood in ... More

Proofs of Vector Identities Using TensorsJun 11 2014The vector algebra and calculus are frequently used in many branches of Physics, for example, classical mechanics, electromagnetic theory, Astrophysics, Spectroscopy, etc. Important vector identities with the help of Levi-Civita symbols and Kronecker ... More

Comparison of Random Waypoint & Random Walk Mobility Model under DSR, AODV & DSDV MANET Routing ProtocolsApr 13 2011Apr 14 2011Mobile Adhoc Network is a kind of wireless ad hoc network where nodes are connected wirelessly and the network is self configuring. MANET may work in a standalone manner or may be a part of another network. In this paper we have compared Random Walk Mobility ... More

A Schreier Domain Type ConditionDec 01 2011We study the integral domains D satisfying the following condition: whenever I >AB with I,A,B nonzero ideals, there exist ideals A'>A and B'>B such that I=A'B'.

Digital Libraries: From Process Modelling to Grid-based Service Oriented ArchitectureFeb 23 2006Graphical Business Process Modelling Languages (BPML) like Role Activity Diagrams (RAD) provide ease and flexibility for modelling business behaviour. However, these languages show limited applicability in terms of enactment over distributed systems paradigms ... More

A Schreier domain type condition IIDec 06 2011For an integral domain D and a star operation * on D, we study the following condition: whenever I>AB with I, A, B nonzero ideals, there exist nonzero ideals H and J such that I*=(HJ)*, H*>A and J*>B.

A Proposed Decision Support System/Expert System for Guiding Fresh Students in Selecting a Faculty in Gomal University, PakistanApr 09 2011Mar 08 2012This paper presents the design and development of a proposed rule based Decision Support System that will help students in selecting the best suitable faculty/major decision while taking admission in Gomal University, Dera Ismail Khan, Pakistan. The basic ... More

A Customized Lattice Reduction Multiprocessor for MIMO DetectionJan 17 2015Lattice reduction (LR) is a preprocessing technique for multiple-input multiple-output (MIMO) symbol detection to achieve better bit error-rate (BER) performance. In this paper, we propose a customized homogeneous multiprocessor for LR. The processor ... More

Effect of Solar-Terrestrial Phenomena on Solar Cell's EfficiencyJun 07 2012It is assumed that the solar cell efficiency of PV device is closely related to the solar irradiance, considered the solar parameter Global Solar Irradiance (G) and the meteorological parameters like daily data of Earth Skin Temperature (E), Average Temperature ... More

Incrementally Learning Functions of the ReturnJul 05 2019Temporal difference methods enable efficient estimation of value functions in reinforcement learning in an incremental fashion, and are of broader interest because they correspond learning as observed in biological systems. Standard value functions correspond ... More

Organizing Experience: A Deeper Look at Replay Mechanisms for Sample-based Planning in Continuous State DomainsJun 12 2018Model-based strategies for control are critical to obtain sample efficient learning. Dyna is a planning paradigm that naturally interleaves learning and planning, by simulating one-step experience to update the action-value function. This elegant planning ... More

Casimir interactions of an object inside a spherical metal shellAug 22 2009Nov 03 2009We investigate the electromagnetic Casimir interactions of an object contained within an otherwise empty, perfectly conducting spherical shell. For a small object we present analytical calculations of the force, which is directed away from the center ... More

An Architecture for Integrated Intelligence in Urban Management using Cloud ComputingFeb 24 2012With the emergence of new methodologies and technologies it has now become possible to manage large amounts of environmental sensing data and apply new integrated computing models to acquire information intelligence. This paper advocates the application ... More

Whistler instability in a semi-relativistic bi-Maxwellian plasmaApr 09 2013Employing linearized Vlasov-Maxwell system, the Weibel instability embedded in an ambient magnetic field is discussed for a semi-relativistic bi-Maxwellian distribution hoping such a scenario occurs in some relativistic environments e.g., gamma-ray burst ... More

Casimir potential of a compact object enclosed by a spherical cavityAug 25 2010Oct 20 2010We study the electromagnetic Casimir interaction of a compact object contained inside a closed cavity of another compact object. We express the interaction energy in terms of the objects' scattering matrices and translation matrices that relate the coordinate ... More

A Semantic Grid-based E-Learning Framework (SELF)Feb 09 2005E-learning can be loosely defined as a wide set of applications and processes, which uses available electronic media (and tools) to deliver vocational education and training. With its increasing recognition as an ubiquitous mode of instruction and interaction ... More

Planning with Expectation ModelsApr 02 2019Distribution and sample models are two popular model choices in model-based reinforcement learning (MBRL). However, learning these models can be intractable, particularly when the state and action spaces are large. Expectation models, on the other hand, ... More

Automatic Vehicle Checking Agent (VCA)Apr 09 2011Dec 03 2011A definition of intelligence is given in terms of performance that can be quantitatively measured. In this study, we have presented a conceptual model of Intelligent Agent System for Automatic Vehicle Checking Agent (VCA). To achieve this goal, we have ... More

Planning with Expectation ModelsApr 02 2019Apr 03 2019Distribution and sample models are two popular model choices in model-based reinforcement learning (MBRL). However, learning these models can be intractable, particularly when the state and action spaces are large. Expectation models, on the other hand, ... More

Optical-approximation analysis of sidewall-spacing effects on the force between two squares with parallel sidewallsSep 05 2007Sep 05 2007Using the ray-optics approximation, we analyze the Casimir force in a two dimensional domain formed by two metallic blocks adjacent to parallel metallic sidewalls, which are separated from the blocks by a finite distance h. For h > 0, the ray-optics approach ... More

Dirac semimetal in three dimensionsNov 28 2011Feb 17 2012In a Dirac semimetal, the conduction and valence bands contact only at discrete (Dirac) points in the Brillouin zone (BZ) and disperse linearly in all directions around these critical points. Including spin, the low energy effective theory around each ... More

Novel two-dimensional Nb$_2$C MXene for highest-Tc superconductivityAug 12 2019Superconductivity in two-dimensional (2D) materials is a hot topic of research today owing to their promising technological applications. The present work reports superconductivity in a new 2D Nb$_2$C MXene with highest critical temperature of 12.5 K ... More

Bulk Dirac points in distorted spinelsSep 23 2013We report on a Dirac-like Fermi surface in three-dimensional bulk materials in a distorted spinel structure on the basis of density functional theory (DFT) as well as tight-binding theory. The four examples we provide in this paper are BiZnSiO4, BiCaSiO4, ... More

Novel highest-Tc superconductivity in two-dimensional Nb2C MXeneAug 12 2019Aug 16 2019Currently, superconductivity in two-dimensional (2D) materials is a hot topic of research owing to their potential technological applications. Here, we report observation of superconductivity in a 2D Nb2C MXene with transition temperature of 12.5 K, which ... More

Spin texture on the Fermi surface of tensile strained HgTeJun 04 2012Dec 19 2012We present ab initio and k.p calculations of the spin texture on the Fermi surface of tensile strained HgTe, which is obtained by stretching the zincblende lattice along the (111) axis. Tensile strained HgTe is a semimetal with pointlike accidental degeneracies ... More

Dose Verification of Volumetric Modulated Arc Therapy using One and Two Dimensional DosimetersMay 17 2018Purpose: To verify dose delivery and quality assurance of volumetric modulated arc therapy (VMAT) for head and neck cancer. Method: The Imaging and Radiation Oncology Core Houston (IROC-H) head and neck phantom with thermo- luminescent dosimeters (TLDs) ... More