Results 1  10
of
333
Optimal distributed online prediction using minibatches
, 2010
"... Online prediction methods are typically presented as serial algorithms running on a single processor. However, in the age of webscale prediction problems, it is increasingly common to encounter situations where a single processor cannot keep up with the high rate at which inputs arrive. In this wor ..."
Abstract

Cited by 69 (7 self)
 Add to MetaCart
Online prediction methods are typically presented as serial algorithms running on a single processor. However, in the age of webscale prediction problems, it is increasingly common to encounter situations where a single processor cannot keep up with the high rate at which inputs arrive. In this work, we present the distributed minibatch algorithm, a method of converting many serial gradientbased online prediction algorithms into distributed algorithms. We prove a regret bound for this method that is asymptotically optimal for smooth convex loss functions and stochastic inputs. Moreover, our analysis explicitly takes into account communication latencies between nodes in the distributed environment. We show how our method can be used to solve the closelyrelated distributed stochastic optimization problem, achieving an asymptotically linear speedup over multiple processors. Finally, we demonstrate the merits of our approach on a webscale online prediction problem.
Study of Relation Between Clinical and Biological Findings in 7 Subjects
"... between clinical and biochemical findings in 7 subjects. Seven subjects with raised plasma histidine and low skin histidase levels (histidinaemia) are described: 4 were severely retarded, 2 showing in addition features of an early infantile psychosis (autism); 3 were of normal intelligence. There we ..."
Abstract
 Add to MetaCart
. The natural history of the condition is further examined, particularly the question of deterioration at time of seizures or infection. In 1961, Ghadimi, Partington, and Hunter described two sisters with a raised plasma histidine level: a 3yearold with marked speech delay but normal nonverbal IQ, and a 4
Optimal stochastic approximation algorithms for strongly convex stochastic composite optimization, II: shrinking procedures and optimal algorithms
, 2010
"... In this paper we present a generic algorithmic framework, namely, the accelerated stochastic approximation (ACSA) algorithm, for solving strongly convex stochastic composite optimization (SCO) problems. While the classical stochastic approximation (SA) algorithms are asymptotically optimal for solv ..."
Abstract

Cited by 50 (8 self)
 Add to MetaCart
In this paper we present a generic algorithmic framework, namely, the accelerated stochastic approximation (ACSA) algorithm, for solving strongly convex stochastic composite optimization (SCO) problems. While the classical stochastic approximation (SA) algorithms are asymptotically optimal for solving differentiable and strongly convex problems, the ACSA algorithm, when employed with proper stepsize policies, can achieve optimal or nearly optimal rates of convergence for solving different classes of SCO problems during a given number of iterations. Moreover, we investigate these ACSA algorithms in more detail, such as, establishing the largedeviation results associated with the convergence rates and introducing efficient validation procedure to check the accuracy of the generated solutions.
SYNTHESIS AND CONTROL OF WHOLEBODY BEHAVIORS IN HUMANOID SYSTEMS
, 2007
"... A great challenge for robotic systems is their ability to carry on complex manipulation and locomotion tasks while responding to the changing environment. To allow robots to operate in human environments there is a strong need to develop new control architectures that can provide advanced task cap ..."
Abstract

Cited by 45 (9 self)
 Add to MetaCart
A great challenge for robotic systems is their ability to carry on complex manipulation and locomotion tasks while responding to the changing environment. To allow robots to operate in human environments there is a strong need to develop new control architectures that can provide advanced task capabilities and interactive skills. These architectures must be effective in coordinating wholebody behaviors for various control objectives while complying with balance stability, contact stance, and other dynamic constraints. In addition, to facilitate the integration of robots in human environments, it is desirable for their motions and task behaviors to be compatible with those of humans. In this thesis, we present a control methodology for the synthesis of realtime wholebody control behaviors in humanoid systems. The work is presented in three parts. First, we establish mathematical foundations that characterize the kinematic and dynamic behaviors of task and postural criteria under balance and contact stability constraints. We identify the dynamic behavior of postural tasks operating in the null space of operational
Block stochastic gradient update method
, 2015
"... Stochastic gradient method Consider the stochastic programming min x∈X F(x) = Eξf (x; ξ). Stochastic gradient update (SG): xk+1 = PX xk − αk g̃k • g̃k a stochastic gradient, often E[g̃k] ∈ ∂F(xk) • Originally for stochastic problem where exact gradient not available • Now also popular for determin ..."
Abstract
 Add to MetaCart
) for strongly convex problem (e.g., [Nemirovski et. al’09]) • For deterministic problem, linear convergence is possible if exact gradient allowed periodically [XiaoZhang’14] • Convergence in terms of firstorder optimality condition for nonconvex problem [GhadimiLan’13] 3 / 26 Block gradient descent Consider
Minimizing Finite Sums with the Stochastic Average Gradient
, 2013
"... We propose the stochastic average gradient (SAG) method for optimizing the sum of a finite number of smooth convex functions. Like stochastic gradient (SG) methods, the SAG method’s iteration cost is independent of the number of terms in the sum. However, by incorporating a memory of previous gradie ..."
Abstract

Cited by 34 (2 self)
 Add to MetaCart
We propose the stochastic average gradient (SAG) method for optimizing the sum of a finite number of smooth convex functions. Like stochastic gradient (SG) methods, the SAG method’s iteration cost is independent of the number of terms in the sum. However, by incorporating a memory of previous gradient values the SAG method achieves a faster convergence rate than blackbox SG methods. The convergence rate is improved from O(1 / √ k) to O(1/k) in general, and when the sum is stronglyconvex the convergence rate is improved from the sublinear O(1/k) to a linear convergence rate of the form O(ρ k) for ρ < 1. Further, in many cases the convergence rate of the new method is also faster than blackbox deterministic gradient methods, in terms of the number of gradient evaluations. Numerical experiments indicate that the new algorithm often dramatically outperforms existing SG and deterministic gradient methods, and that the performance may be further improved through the use of nonuniform sampling strategies. 1
Stochastic Alternating Direction Method of Multipliers
"... The Alternating Direction Method of Multipliers (ADMM) has received lots of attention recently due to the tremendous demand from largescale and datadistributed machine learning applications. In this paper, we present a stochastic setting for optimization problems with nonsmooth composite objectiv ..."
Abstract

Cited by 30 (0 self)
 Add to MetaCart
The Alternating Direction Method of Multipliers (ADMM) has received lots of attention recently due to the tremendous demand from largescale and datadistributed machine learning applications. In this paper, we present a stochastic setting for optimization problems with nonsmooth composite objective functions. To solve this problem, we propose a stochastic ADMM algorithm. Our algorithm applies to a more general class of convex and nonsmooth objective functions, beyond the smooth and separable least squares loss used in lasso. We also demonstrate the rates of convergence for our algorithm under various structural assumptions of the stochastic function: O(1 / √ t) for convex functions and O(log t/t) for strongly convex functions. Compared to previous literature, we establish the convergence rate of ADMM for convex problems in terms of both the objective value and the feasibility violation. A novel application named GraphGuided SVM is proposed to demonstrate the usefulness of our algorithm.
par ARNAUD Ophélie ETUDE FONCTIONNELLE DE LA REGION INTRACELLULAIRE D’ABCG2 ET MODULATION D’ABCG2 ET ABCB1 HUMAINS PAR DES PEPTIDOMIMETIQUES NON COMPETITIFS
, 2013
"... M le docteur Attilio DiPietro JURY: M le docteur JeanMichel Jault M le docteur Stéphane Orlowski M le professeur Charles Dumontet Mme le professeur. Joëlle Paris M le docteur. Pierre Falson M le docteur. Attilio DiPietro ..."
Abstract
 Add to MetaCart
M le docteur Attilio DiPietro JURY: M le docteur JeanMichel Jault M le docteur Stéphane Orlowski M le professeur Charles Dumontet Mme le professeur. Joëlle Paris M le docteur. Pierre Falson M le docteur. Attilio DiPietro
Computation of a Singlephase ShellType Transformer Windings Forces Caused by Inrush and Shortcircuit Currents 1 M.B.B. Sharifian,
"... Abstract: This research studies the forces on the windings of transformer due to inrush current. These forces are compared with the corresponding forces due to shortcircuit of the windings. Twodimensional finite element computation of a singlephase shelltype transformer is carried out based on th ..."
Abstract
 Add to MetaCart
Abstract: This research studies the forces on the windings of transformer due to inrush current. These forces are compared with the corresponding forces due to shortcircuit of the windings. Twodimensional finite element computation of a singlephase shelltype transformer is carried out based on the maximum permissible inrush current value where its amplitude is the same as the rated shortcircuit current. To verify the computation results, they are compared with those recently obtained using Artificial Neural Network (ANN). Key words: Inrush current, transformer, short circuit, finite element method
Results 1  10
of
333