Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 68 for “"Gradient Estimation"”.
-
Gradient Estimation for Attractor Networks
… this hypothesis in several ways. We study the gradient estimation and optimization procedures for several variants of these networks. We show how the convergence of the gradient estimation procedures are related to the properties of the networks. Then we consider how to tune the relative rates …
-
Proxy Gradient Estimation for Black-Box Feature Attribution
… can exploit full model details, such as exact gradient measurements. To bridge the performance gap between black-box and white-box feature attribution methods, this dissertation introduces a novel framework based on gradient estimation, which is then extended to a more general formulation …
-
Models Of Disparity Gradient Estimation In The Visual Cortex
… oriented in such way have non-zero disparity gradient. The processes that take place in the brain to recover disparity gradient and to create the correct perception of slant in depth are not well understood. Changes in local disparity are an obvious disparity gradient cue. Here, we consider an …
-
Gradient estimation for discrete variables via dependent Monte Carlo samples
… tasks, and we often need to backpropagate the gradient through them. One domain is variational inference, where training discrete latent variable models requires gradient estimates of a high dimensional discrete distribution because we are backpropagating through discrete stochastic layer in a …
-
Gradient Estimation and Variance Reduction in Stochastic and Deterministic Models
… fields. For all of these types of models, gradient-based optimization remains as the dominant paradigm for model fitting, control, and more. This dissertation considers unconstrained, nonlinear optimization problems, with a focus on the gradient itself, that key quantity which enables the …
-
Stochastic Control and Stochastic Gradient Estimation with Applications to Kidney Transplantation
… methods in stochastic control and stochastic gradient estimation, with applications to kidney transplantation. The first part studies the optimal acceptance of incompatible kidneys for individual patients with end-stage kidney disease. Modern desensitization techniques allow patients to accept …
-
Energy Density Gradient Estimation for Monte Carlo Methods in Mix Models
… representation is derived to approximate the gradient of the energy density, which is the coupling quantity between the radiation and hydrodynamics solve. Results using this finite element estimator, are presented for two problems of different complexity, and compared to results using a finite …
-
Use of Phase and Amplitude Gradient Estimation for Acoustic Source Characterization and Localization
… sources. Recently, the phase and amplitude gradient estimator (PAGE) method has been developed to reduce error and extend bandwidth of energy-based quantity estimates. To inform uses and applications of the method, analytical and experimental characterizations of the method are presented. …
-
Characterization of gradient estimators for stochastic activity networks
… properties of Monte Carlo simulation-based gradient estimation techniques for performance measures in stochastic activity networks (SANs) using the estimators' variance as the comparison criterion. When analyzing SANs, both performance measures and their sensitivities (gradient, Hessian) are …
-
Optimal Variance Swaps Portfolios and Estimating Greeks for Variance-Gamma
… optimal variance swap portfolio. By implementing gradient estimation techniques, we estimate the Greeks for a series of basket options called Mountain Range options. Constructing an optimal variance swap portfolio consists of two steps: evaluations and optimization. Each variance swap has two …
-
Differentiable Monte Carlo Samplers with Piecewise Deterministic Markov Processes
Gradient estimation by Monte Carlo methods, to e.g. find optimization directions, is an important component of many problems in statistics and machine learning. In one approach, related to the reparameterization trick, the sampling method itself is differentiated pathwise to obtain a sampler for …
-
Automatic Differentiation for Climate Model Calibration: A Gradient-Based Framework for Parameter Estimation in SpeedyWeather.jl
… via automatic differentiation (AD) can provide gradients for systematic, transparent, and scalable calibration, but chaotic dynamics make long-term gradients unstable and prohibitively expensive. This thesis develops a practical workflow to overcome this barrier. Using the Lorenz-63 system, we …
-
Analog Signal Processor for Adaptive Antenna Arrays
… The circuit performs continuous-time gradient descent using simultaneous perturbation gradient estimation. Simulations were performed using Agilent ADS circuit simulator. Field tests were performed in a realistic scenario using a satellite signal. The results were comparable to the …
-
Entropy estimation and optimization and their applications in Bayesian experimental design
… to this thesis is the challenge of entropy estimation and optimization, particularly in high-dimensional spaces, and their applications in Bayesian Experimental Design (BED). The thesis presents and addresses three primary research objectives. The first is a comprehensive survey of existing …
-
Locality Sensitive Sampling for Extreme-Scale Optimization and Deep Learning
… (LSH) as an adaptive sampler for large-scale estimations. We first introduce the chicken-and-the-egg loop problem in large-scale optimization algorithms. SGD estimates the gradient by uniform sampling with sample size one. There have been several other works that suggest faster epoch-wise …
-
How to descend a rocky slope : numerical techniques for the solution of noisy optimization problems
… Filtering (NNAIF), and Linear Model Based Gradient Estimation (LMBGE). SP-BFGS is a new variant of the Broyden-Fletcher-Goldfarb-Shanno (BFGS) quasi-Newton method designed to perform well when gradient measurements are corrupted by noise. SP-BFGS provides a means of incrementally updating …
-
Between integrals and optima: new methods for scalable machine learning
… scalable computational methods, like stochastic gradient descent or Monte Carlo methods, that undergird learning algorithms. This thesis contributes four new scalable methods for distinct problems that arise in machine learning. It introduces a new method for gradient estimation in discrete …
-
Towards implementing deep learning in analog hardware using forward mode predictive coding
… integrates predictive coding and forward mode gradient estimation to enable parallelizable local learning in neural networks – designed specifically for analog hardware implementation. Unlike backpropagation, which is optimized for digital architectures and relies on partial derivative …
-
Stochastic Simulation: New Stochastic Approximation Methods and Sensitivity Analyses
… study the sensitivity of SA and of a stochastic gradient method to various input parameters. First, we summarize the most common stochastic gradient estimation techniques, both direct and indirect, as well as the two classical SA algorithms, Robbins-Monro (RM) and Kiefer-Wolfowitz (KW), followed …
Page 1 of 4