Sequential monte carlo methods To train neural network models by , d, M, M, Gee, A, Doucet, A

“…We discuss a novel strategy for training neural networks using sequential Monte Carlo algorithms and propose a new hybrid gradient descent sampling importance resampling algorithm (HySIR). …”

Mean field analysis of deep neural networks by Sirignano, J, Spiliopoulos, K

“…We analyze multilayer neural networks in the asymptotic regime of simultaneously (a) large network sizes and (b) large numbers of stochastic gradient descent training iterations. We rigorously establish the limiting behavior of the multilayer neural network output. …”

Tracking-based distributed equilibrium seeking for aggregative games by Carnevale, G, Fabiani, F, Fele, F, Margellos, K, Notarstefano, G

“…We first consider the case where local constraints are present and we design an algorithm combining, for each agent, (i) the projected pseudo-gradient descent and (ii) a tracking mechanism to locally reconstruct the aggregative variable. …”

Sex-specific genetic data supports one of two alternative versions of the foundation of the ruling dynasty of the Nso in Cameroon by Veeramah, K, Zeitlyn, D, Fanso, V, Mendell, N, Connell, B, Weale, M, Bradman, N, Thomas, M

“…Sex-specific genetic data favor a specific variant of the oral history of the kingdom of Nso (a Grassfields city-state in Cameroon) in which the royal family traces its descent from a founding ancestress who married into an autochthonous hunter-gatherer group. …”

Fully-convolutional Siamese networks for object tracking by Bertinetto, L, Valmadre, J, Henriques, JF, Vedaldi, A, Torr, PHS

“…However, when the object to track is not known beforehand, it is necessary to perform Stochastic Gradient Descent online to adapt the weights of the network, severely compromising the speed of the system. …”

Global rates of convergence for nonconvex optimization on manifolds by Boumal, N, Absil, P, Cartis, C

“…We consider the minimization of a cost function $f$ on a manifold $M$ using Riemannian gradient descent and Riemannian trust regions (RTR). We focus on satisfying necessary optimality conditions within a tolerance $\varepsilon$. …”

An approximate sampling formula under genetic hitchhiking by Etheridge, A, Pfaffelhuber, P, Wakolbinger, A

“…In particular we see that, with this order of accuracy, in a sample of fixed size there are at most two nonsingleton families of individuals which are identical by descent at the neutral locus from the beginning of the sweep. …”

Joint training of generic CNN-CRF models with stochastic optimization by Kirillov, A, Schlesinger, D, Zheng, S, Savchynskyy, B, Torr, P, Rother, C

“…While stochastic gradient descent is a standard technique for CNN training, it was not used for joint models so far. …”

Large-scale association analysis provides insights into the genetic architecture and pathophysiology of type 2 diabetes. by Morris, A, Voight, B, Teslovich, T, Ferreira, T, Segrè, A, Steinthorsdottir, V, Strawbridge, R, Khan, H, Grallert, H, Mahajan, A, Prokopenko, I, Kang, H, Dina, C, Esko, T, Fraser, R, Kanoni, S, Kumar, A, Lagou, V, Langenberg, C, Luan, J, Lindgren, C, Müller-Nurasyid, M, Pechlivanis, S, Rayner, N, Scott, L

“…To extend understanding of the genetic architecture and molecular basis of type 2 diabetes (T2D), we conducted a meta-analysis of genetic variants on the Metabochip, including 34,840 cases and 114,981 controls, overwhelmingly of European descent. We identified ten previously unreported T2D susceptibility loci, including two showing sex-differentiated association. …”

Segmentation on surfaces with the closest point method by Tian, L, Macdonald, C, Ruuth, S

“…This leads to gradient descent equations which are level set equations on surfaces. …”

Global convergence analysis of a new nonlinear conjugate gradient coefficient with strong wolfe line search by Abdelrahman Abdalla, A., Mamat, M., Rivaie, M., Omer, O.

“…The global convergence and sufficient descent property is established using inexact line searches to determine that α k is the step size of CG methods. …”
Get full text

Nonlinear control with linearized models and neural networks by Hussain, Mohd Azlan, Allwright, J.C., Kershenbaum, L.S.

“…Online adaptation of the network is performed using steepest descent with a dead zone function. The proposed strategy is applied to two case studies for output tracking of set points. …”

A harmony search-based learning algorithm for epileptic seizure prediction by Kee, Huong Lai, Zainuddin, Zarita, Ong, Pauline

“…Apart from the traditional gradient descent-based approach, metaheuristic algorithms can also be used to determine these parameters. …”
Get full text

Trust-region based methods for unconstrained global optimization by Kerk, Lee Chang

“…The developed methods have been proven to satisfy descent property, global convergence and low time complexities. …”
Get full text

Algorithm for solution of non-convex optimization problem through piece-wise convex transformation by Ahmad, Rohanin, Kerk, Lee Chang

“…The identified convex parts are called trusted intervals. The descent property and the global convergence of the method was shown in this paper. 15 test problems have been used to show the ability of the algorithm proposed in locating global minimizer.…”

Sensitivity of Three Patient-Specific Quality Assurance Systems to MLC Aperture Errors With Volumetric Modulated Arc Therapy by Jinyou Hu MS, Shaoxian Gu MS, Ningyu Wang MS, Fengjie Cui MS, Shengyuan Zhang MS, Chuou Yin MS, Yunzhu Cai MS, Chengjun Gou PhD, Lian Zou PhD, Zhangwen Wu PhD

“…The descent gradient of PD was larger than AC, except for the case at the 2%/2mm criteria. …”
Get full text

COMPARISON AND ANALYSIS OF NONLINEAR LEAST SQUARES METHODS FOR VISION BASED NAVIGATION (VBN) ALGORITHMS by B. Sheta, M. Elhabiby, M. Elhabiby, N. Sheimy

“…As for the Levenberg-Marquardt method, which is considered as a modified Gauss-Newton algorithm employing the trust region approach where a scalar is introduced to assess the choice of the magnitude and the direction of the descent. This scalar determines whether the Gauss-Newton method direction or the steepest descent method direction will be used as an adaptive approach for both linear and non-linear mathematical models and it successfully converged and achieved the relative optimum solution. …”
Get full text

Unsupervised Sim-to-Real Adaptation for Environmental Recognition in Assistive Walking by Chuheng Chen, Kuangen Zhang, Yuquan Leng, Xinxing Chen, Chenglong Fu

“…To mitigate this gap, this paper presents an unsupervised sim-to-real adaptation method to accurately classify five common real-world (level ground, stair ascent, stair descent, ramp ascent and ramp descent) and assist amputee’s terrain-adaptive locomotion. …”
Get full text

Fundamental Limits of Learning for Generalizability, Data Resilience, and Resource Efficiency by Blanchard, Moïse

“…In exponential-accuracy regimes, these algorithms strictly improve the oracle-complexity of gradient descent while preserving the same optimal memory usage. …”
Get full text

Identification of risk loci and a polygenic risk score for lung cancer: a large-scale prospective cohort study in Chinese populations by Dai, J, Lv, J, Zhu, M, Wang, Y, Qin, N, Ma, H, He, Y, Zhang, R, Tan, W, Fan, J, Wang, T, Zheng, H, Sun, Q, Wang, L, Huang, M, Ge, Z, Yu, C, Guo, Y, Wang, T, Wang, J, Xu, L, Wu, W, Chen, L, Bian, Z, Walters, R, Millwood, I, Li, X, Wang, X, Hung, R, Christiani, D, Chen, H, Wang, M, Wang, C, Jiang, Y, Chen, K, Chen, Z, Jin, G, Wu, T, Lin, D, Hu, Z, Amos, C, Wu, C, Wei, Q, Jia, W, Li, L, Shen, H

“…<p><strong>Methods</strong></p> To systematically identify genetic variants for NSCLC risk, we newly genotyped 19 546 samples from Chinese NSCLC cases and controls from the Nanjing Medical University Global Screening Array Project and did a meta-analysis of genome-wide association studies (GWASs) of 27 120 individuals with NSCLC and 27 355 without NSCLC (13 327 cases and 13 328 controls of Chinese descent as well as 13 793 cases and 14 027 controls of European descent). …”