Publications related to Is there an analog of Nesterov acceleration for gradient-based MCMC?

On the performance of random reshuffling in stochastic learning

Ali H. Sayed, Stefan Vlaski, Bicheng Ying, Kun Yuan

In empirical risk optimization, it has been observed that gradient descent implementations that rely on random reshuffling of the data achieve better performance than implementations that rely on sampling the data randomly and independently of each other. ...

2017

Fréchet means in Wasserstein space

Yoav Zemel

This work studies the problem of statistical inference for Fréchet means in the Wasserstein space of measures on Euclidean spaces,

\mathcal W_2 ( \mathbb R^d )

. This question arises naturally from the problem of separating amplitude and phase variation i ...

EPFL2017

Attributing Aliasing Artifacts in Dirty Intensity Fields to their Parent Source

Matthieu Martin Jean-André Simeoni

Estimating intensity fields of stochastic phenomenons is of crucial interest in many scientific applications. Typical experimental setups involve an acquisition system, that subsequently filters and samples the probed intensity field. This equivalently def ...

2016

Stochastic gradient descent with finite samples sizes

Ali H. Sayed, Stefan Vlaski, Bicheng Ying, Kun Yuan

The minimization of empirical risks over finite sample sizes is an important problem in large-scale machine learning. A variety of algorithms has been proposed in the literature to alleviate the computational burden per iteration at the expense of converge ...

IEEE2016

MATHICSE Technical Report : Discrete least-squares approximations over optimized downward closed polynomial spaces in arbitrary dimension

Fabio Nobile, Giovanni Migliorati

We analyze the accuracy of the discrete least-squares approximation of a function u in multivariate polynomial spaces

P_\Lambda := span\{y \mapsto y^\nu | \nu \in \Lambda\}

with

\Lambda\subset N_0^d

over the domain

\Gamma := [-1,1]^d

, based on the sa ...

MATHICSE2015

Discrete least squares polynomial approximation with random evaluations − application to parametric and stochastic elliptic PDEs

Fabio Nobile, Giovanni Migliorati

Motivated by the numerical treatment of parametric and stochastic PDEs, we analyze the least-squares method for polynomial approximation of multivariate functions based on random sampling according to a given probability measure. Recent work has shown that ...

2015

Stochastic Spectral Descent for Restricted Boltzmann Machines.

Volkan Cevher

Restricted Boltzmann Machines (RBMs) are widely used as building blocks for deep learning models. Learning typically proceeds by using stochastic gradient descent, and the gradients are estimated with sampling methods. However, the gradient estimation is a ...

2015

Existence and Uniqueness of Maximal Regular Flows for Non-smooth Vector Fields

Maria Colombo

In this paper we provide a complete analogy between the Cauchy-Lipschitz and the DiPerna-Lions theories for ODE's, by developing a local version of the DiPerna-Lions theory. More precisely, we prove the existence and uniqueness of a maximal regular flow fo ...

2015

Electrical Neuroimaging with Irrotational Sources

This paper discusses theoretical aspects of the modeling of the sources of the EEG (i.e., the bioelectromagnetic inverse problem or source localization problem). Using theHelmholtz decomposition (HD) of the current density vector (CDV) of the primary curre ...

Hindawi Publishing Corporation2015

Analytic Sensing For Multi-Layer Spherical Models With Application To EEG Source Imaging

Dimitri Nestor Alice Van De Ville, Thierry Blu, Djano Kandaswamy

Source imaging maps back boundary measurements to underlying generators within the domain; e. g., retrieving the parameters of the generating dipoles from electrical potential measurements on the scalp such as in electroencephalography (EEG). Fitting such ...

Amer Inst Mathematical Sciences2013

Is there an analog of Nesterov acceleration for gradient-based MCMC?

Graph Chatbot

Chat with Graph Search

On the performance of random reshuffling in stochastic learning

Fréchet means in Wasserstein space

Attributing Aliasing Artifacts in Dirty Intensity Fields to their Parent Source

Stochastic gradient descent with finite samples sizes

MATHICSE Technical Report : Discrete least-squares approximations over optimized downward closed polynomial spaces in arbitrary dimension

Discrete least squares polynomial approximation with random evaluations − application to parametric and stochastic elliptic PDEs

Stochastic Spectral Descent for Restricted Boltzmann Machines.

Existence and Uniqueness of Maximal Regular Flows for Non-smooth Vector Fields

Electrical Neuroimaging with Irrotational Sources

Analytic Sensing For Multi-Layer Spherical Models With Application To EEG Source Imaging

Stochastic Spectral Descent for Restricted Boltzmann Machines.

Fréchet means in Wasserstein space

Attributing Aliasing Artifacts in Dirty Intensity Fields to their Parent Source

Discrete least squares polynomial approximation with random evaluations − application to parametric and stochastic elliptic PDEs

Analytic Sensing For Multi-Layer Spherical Models With Application To EEG Source Imaging

Existence and Uniqueness of Maximal Regular Flows for Non-smooth Vector Fields

Electrical Neuroimaging with Irrotational Sources

Stochastic gradient descent with finite samples sizes

On the performance of random reshuffling in stochastic learning

MATHICSE Technical Report : Discrete least-squares approximations over optimized downward closed polynomial spaces in arbitrary dimension