Stochastic Discrete Clenshaw-Curtis Quadrature

10/26/2020
by   Katharina Morik, et al.
0

The partition function is fundamental for probabilistic graphical models—it is required for inference, parameter estimation, and model selection. Evaluating this function corresponds to discrete integration, namely a weighted sum over an exponentially large set. This task quickly becomes intractable as the dimensionality of the problem increases. We propose an approximation scheme that, for any discrete graphical model whose parameter vector has bounded norm, estimates the partition function with arbitrarily small error. Our algorithm relies on a near minimax optimal polynomial approximation to the potential function and a Clenshaw-Curtis style quadrature. Furthermore, we show that this algorithm can be randomized to split the computation into a high-complexity part and a low-complexity part, where the latter may be carried out on small computational devices. Experiments confirm that the new randomized algorithm is highly accurate if the parameter norm is small, and is otherwise comparable to methods with unbounded error.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/27/2013

Taming the Curse of Dimensionality: Discrete Integration by Hashing and Optimization

Integration is affected by the curse of dimensionality and quickly becom...
research
09/12/2014

Hardness of parameter estimation in graphical models

We consider the problem of learning the canonical parameters specifying ...
research
01/11/2017

Guaranteed Parameter Estimation for Discrete Energy Minimization

Structural learning, a method to estimate the parameters for discrete en...
research
02/14/2012

Sum-Product Networks: A New Deep Architecture

The key limiting factor in graphical model inference and learning is the...
research
06/23/2023

Randomized Complexity of Parametric Integration and the Role of Adaption II. Sobolev Spaces

We study the complexity of randomized computation of integrals depending...
research
06/13/2017

Lost Relatives of the Gumbel Trick

The Gumbel trick is a method to sample from a discrete probability distr...
research
10/13/2019

AdaWISH: Faster Discrete Integration via Adaptive Quantiles

Discrete integration in a high dimensional space of n variables poses fu...

Please sign up or login with your details

Forgot password? Click here to reset