arXiv subjects

Simo Särkkä

Publications and source records attributed to Simo Särkkä.

At least 55 records · Page 3Linked to original sources

arXiv2023

Gaussian-Based Parametric Bijections For Automatic Projection Filters

The automatic projection filter is a recently developed numerical method for projection filtering that leverages sparse-grid integration and automatic differentiation. However, its accuracy is highly sensitive to the accuracy of the cumulant-generating function computed via the sparse-grid integration, which in turn is also sensitive to the choice of the bijection from the canonical hypercube to the state space. In this paper, we propose two new adaptive parametric bijections for the automatic projection filter. The first bijection relies on the minimization of Kullback--Leibler divergence, whereas the second method employs the sparse-grid Gauss--Hermite quadrature. The two new bijections allow the sparse-grid nodes to adaptively move within the high-density region of the state space, resulting in a substantially improved approximation while using only a small number of quadrature nodes. The practical applicability of the methodology is illustrated in three simulated nonlinear filtering problems.

math.OC↗

arXiv2023

A probabilistic Taylor expansion with Gaussian processes

We study a class of Gaussian processes for which the posterior mean, for a particular choice of data, replicates a truncated Taylor expansion of any order. The data consist of derivative evaluations at the expansion point and the prior covariance kernel belongs to the class of Taylor kernels, which can be written in a certain power series form. We discuss and prove some results on maximum likelihood estimation of parameters of Taylor kernels. The proposed framework is a special case of Gaussian process regression based on data that is orthogonal in the reproducing kernel Hilbert space of the covariance kernel.

cs.LG↗

arXiv2023

Single Qubit State Estimation on NISQ Devices with Limited Resources and SIC-POVMs

Current quantum computers have the potential to overcome classical computational methods, however, the capability of the algorithms that can be executed on noisy intermediate-scale quantum devices is limited due to hardware imperfections. Estimating the state of a qubit is often needed in different quantum protocols, due to the lack of direct measurements. In this paper, we consider the problem of estimating the quantum state of a qubit in a quantum processing unit without conducting direct measurements of it. We consider a parameterized measurement model to estimate the quantum state, represented as a quantum circuit, which is optimized using the quantum tomographic transfer function. We implement and test the circuit using the quantum computer of the Technical Research Centre of Finland as well as an IBM quantum computer. We demonstrate that the set of positive operator-valued measurements used for the estimation is symmetric and informationally complete. Moreover, the resources needed for qubit estimation are reduced when direct measurements are allowed, keeping the symmetric property of the measurements.

quant-ph↗

arXiv2023

A Recursive Newton Method for Smoothing in Nonlinear State Space Models

In this paper, we use the optimization formulation of nonlinear Kalman filtering and smoothing problems to develop second-order variants of iterated Kalman smoother (IKS) methods. We show that Newton's method corresponds to a recursion over affine smoothing problems on a modified state-space model augmented by a pseudo measurement. The first and second derivatives required in this approach can be efficiently computed with widely available automatic differentiation tools. Furthermore, we show how to incorporate line-search and trust-region strategies into the proposed second-order IKS algorithm in order to regularize updates between iterations. Finally, we provide numerical examples to demonstrate the method's efficiency in terms of runtime compared to its batch counterpart.

eess.SP↗

arXiv2023

Fast Dynamic Programming in Trees in the MPC Model

We present a deterministic algorithm for solving a wide range of dynamic programming problems in trees in $O(\log D)$ rounds in the massively parallel computation model (MPC), with $O(n^δ)$ words of local memory per machine, for any given constant $0 < δ< 1$. Here $D$ is the diameter of the tree and $n$ is the number of nodes--we emphasize that our running time is independent of $n$. Our algorithm can solve many classical graph optimization problems such as maximum weight independent set, maximum weight matching, minimum weight dominating set, and minimum weight vertex cover. It can also be used to solve many accumulation tasks in which some aggregate information is propagated upwards or downwards in the tree--this includes, for example, computing the sum, minimum, or maximum of the input labels in each subtree, as well as many inference tasks commonly solved with belief propagation. Our algorithm can also solve any locally checkable labeling problem (LCLs) in trees. Our algorithm works for any reasonable representation of the input tree; for example, the tree can be represented as a list of edges or as a string with nested parentheses or tags. The running time of $O(\log D)$ rounds is also known to be necessary, assuming the widely-believed $2$-cycle conjecture. Our algorithm strictly improves on two prior algorithms: (i) Bateni, Behnezhad, Derakhshan, Hajiaghayi, and Mirrokni [ICALP'18] solve problems of these flavors in $O(\log n)$ rounds, while our algorithm is much faster in low-diameter trees. Furthermore, their algorithm also uses randomness, while our algorithm is deterministic. (ii) Balliu, Latypov, Maus, Olivetti, and Uitto [SODA'23] solve only locally checkable labeling problems in $O(\log D)$ rounds, while our algorithm can be applied to a much broader family of problems.

cs.DC↗

arXiv2023

Parallel square-root statistical linear regression for inference in nonlinear state space models

In this article, we introduce parallel-in-time methods for state and parameter estimation in general nonlinear non-Gaussian state-space models using the statistical linear regression and the iterated statistical posterior linearization paradigms. We also reformulate the proposed methods in a square-root form, resulting in improved numerical stability while preserving the parallelization capabilities. We then leverage the fixed-point structure of our methods to perform likelihood-based parameter estimation in logarithmic time with respect to the number of observations. Finally, we demonstrate the practical performance of the methodology with numerical experiments run on a graphics processing unit (GPU).

stat.CO↗

arXiv2023

On the Convergence of Numerical Integration as a Finite Matrix Approximation to Multiplication Operator

We study the convergence of a family of numerical integration methods where the numerical integral is formulated as a finite matrix approximation to a multiplication operator. For bounded functions, the convergence has already been established using the theory of strong operator convergence. In this article, we consider unbounded functions and domains which pose several difficulties compared to the bounded case. A natural choice of method for this study is the theory of strong resolvent convergence which has previously been mostly applied to study the convergence of approximations of differential operators. The existing theory already includes convergence theorems that can be used as proofs as such for a limited class of functions and extended for wider class of functions in terms of function growth or discontinuity. The extended results apply to all self-adjoint operators, not just multiplication operators. We also show how Jensen's operator inequality can be used to analyse the convergence of an improper numerical integral of a function bounded by an operator convex function.

math.NA↗

arXiv2023

Probabilistic Estimation of Instantaneous Frequencies of Chirp Signals

We present a continuous-time probabilistic approach for estimating the chirp signal and its instantaneous frequency function when the true forms of these functions are not accessible. Our model represents these functions by non-linearly cascaded Gaussian processes represented as non-linear stochastic differential equations. The posterior distribution of the functions is then estimated with stochastic filters and smoothers. We compute a (posterior) Cramér--Rao lower bound for the Gaussian process model, and derive a theoretical upper bound for the estimation error in the mean squared sense. The experiments show that the proposed method outperforms a number of state-of-the-art methods on a synthetic data. We also show that the method works out-of-the-box for two real-world datasets.

stat.ML↗

arXiv2022

Bayes-Newton Methods for Approximate Bayesian Inference with PSD Guarantees

We formulate natural gradient variational inference (VI), expectation propagation (EP), and posterior linearisation (PL) as extensions of Newton's method for optimising the parameters of a Bayesian posterior distribution. This viewpoint explicitly casts inference algorithms under the framework of numerical optimisation. We show that common approximations to Newton's method from the optimisation literature, namely Gauss-Newton and quasi-Newton methods (e.g., the BFGS algorithm), are still valid under this 'Bayes-Newton' framework. This leads to a suite of novel algorithms which are guaranteed to result in positive semi-definite (PSD) covariance matrices, unlike standard VI and EP. Our unifying viewpoint provides new insights into the connections between various inference schemes. All the presented methods apply to any model with a Gaussian prior and non-conjugate likelihood, which we demonstrate with (sparse) Gaussian processes and state space models.

stat.ML↗

arXiv2022

Fourier--Hermite Dynamic Programming for Optimal Control

In this paper, we propose a novel computational method for solving non-linear optimal control problems. The method is based on the use of Fourier--Hermite series for approximating the action-value function arising in dynamic programming instead of the conventional Taylor series expansion used in differential dynamic programming (DDP). The coefficients of the Fourier--Hermite series can be numerically computed by using sigma-point methods, which leads to a novel class of sigma-point based dynamic programming methods. We also prove the quadratic convergence of the method and experimentally test its performance against other methods.

math.OC↗

arXiv2022

Multidimensional Projection Filters via Automatic Differentiation and Sparse-Grid Integration

The projection filter is a technique for approximating the solutions of optimal filtering problems. In projection filters, the Kushner--Stratonovich stochastic partial differential equation that governs the propagation of the optimal filtering density is projected to a manifold of parametric densities, resulting in a finite-dimensional stochastic differential equation. Despite the fact that projection filters are capable of representing complicated probability densities, their current implementations are limited to Gaussian family or unidimensional filtering applications. This work considers a combination of numerical integration and automatic differentiation to construct projection filter algorithms for more generic problems. Specifically, we provide a detailed exposition of this combination for the manifold of the exponential family, and show how to apply the projection filter to multidimensional cases. We demonstrate numerically that based on comparison to a finite-difference solution to the Kushner--Stratonovich equation and a bootstrap particle filter with systematic resampling, the proposed algorithm retains an accurate approximation of the filtering density while requiring a comparatively low number of quadrature points. Due to the sparse-grid integration and automatic differentiation used to calculate the expected values of the natural statistics and the Fisher metric, the proposed filtering algorithms are highly scalable. They therefore are suitable to many applications in which the number of dimensions exceeds the practical limit of particle filters, but where the Gaussian-approximations are deemed unsatisfactory.

math.OC↗

arXiv2022

Online Pole Segmentation on Range Images for Long-term LiDAR Localization in Urban Environments

Robust and accurate localization is a basic requirement for mobile autonomous systems. Pole-like objects, such as traffic signs, poles, and lamps are frequently used landmarks for localization in urban environments due to their local distinctiveness and long-term stability. In this paper, we present a novel, accurate, and fast pole extraction approach based on geometric features that runs online and has little computational demands. Our method performs all computations directly on range images generated from 3D LiDAR scans, which avoids processing 3D point clouds explicitly and enables fast pole extraction for each scan. We further use the extracted poles as pseudo labels to train a deep neural network for online range image-based pole segmentation. We test both our geometric and learning-based pole extraction methods for localization on different datasets with different LiDAR scanners, routes, and seasonal changes. The experimental results show that our methods outperform other state-of-the-art approaches. Moreover, boosted with pseudo pole labels extracted from multiple datasets, our learning-based method can run across different datasets and achieve even better localization results compared to our geometry-based method. We released our pole datasets to the public for evaluating the performance of pole extractors, as well as the implementation of our approach.

cs.RO↗

arXiv2022

The Coupled Rejection Sampler

We propose a coupled rejection-sampling method for sampling from couplings of arbitrary distributions. The method relies on accepting or rejecting coupled samples coming from dominating marginals. Contrary to existing acceptance-rejection coupling methods, the variance of the execution time of the proposed method is limited and stays finite as the two target marginals approach each other in the sense of the total variation norm. In the important special case of coupling multivariate Gaussians with different means and covariances, we derive positive lower bounds for the resulting coupling probability of our algorithm, and we then show how the coupling method can be optimized in closed form. Finally, we show how we can modify the coupled rejection-sampling method to propose from coupled ensemble of proposals, so as to asymptotically recover a maximal coupling. We then apply the method to the problem of coupling rare events samplers, derive a parallel coupled resampling algorithm to use in particle filtering, and show how the coupled rejection-sampler can be used to speed up unbiased MCMC methods based on couplings.

stat.ME↗

arXiv2022

De-Sequentialized Monte Carlo: a parallel-in-time particle smoother

Particle smoothers are SMC (Sequential Monte Carlo) algorithms designed to approximate the joint distribution of the states given observations from a state-space model. We propose dSMC (de-Sequentialized Monte Carlo), a new particle smoother that is able to process $T$ observations in $\mathcal{O}(\log T)$ time on parallel architecture. This compares favourably with standard particle smoothers, the complexity of which is linear in $T$. We derive $\mathcal{L}_p$ convergence results for dSMC, with an explicit upper bound, polynomial in $T$. We then discuss how to reduce the variance of the smoothing estimates computed by dSMC by (i) designing good proposal distributions for sampling the particles at the initialization of the algorithm, as well as by (ii) using lazy resampling to increase the number of particles used in dSMC. Finally, we design a particle Gibbs sampler based on dSMC, which is able to perform parameter inference in a state-space model at a $\mathcal{O}(\log(T))$ cost on parallel hardware.

stat.CO↗

arXiv2022

Uncertainty-aware deep learning methods for robust diabetic retinopathy classification

Automatic classification of diabetic retinopathy from retinal images has been widely studied using deep neural networks with impressive results. However, there is a clinical need for estimation of the uncertainty in the classifications, a shortcoming of modern neural networks. Recently, approximate Bayesian deep learning methods have been proposed for the task but the studies have only considered the binary referable/non-referable diabetic retinopathy classification applied to benchmark datasets. We present novel results by systematically investigating a clinical dataset and a clinically relevant 5-class classification scheme, in addition to benchmark datasets and the binary classification scheme. Moreover, we derive a connection between uncertainty measures and classifier risk, from which we develop a new uncertainty measure. We observe that the previously proposed entropy-based uncertainty measure generalizes to the clinical dataset on the binary classification scheme but not on the 5-class scheme, whereas our new uncertainty measure generalizes to the latter case.

cs.CV↗

arXiv2022

System identification using Bayesian neural networks with nonparametric noise models

System identification is of special interest in science and engineering. This article is concerned with a system identification problem arising in stochastic dynamic systems, where the aim is to estimate the parameters of a system along with its unknown noise processes. In particular, we propose a Bayesian nonparametric approach for system identification in discrete time nonlinear random dynamical systems assuming only the order of the Markov process is known. The proposed method replaces the assumption of Gaussian distributed error components with a highly flexible family of probability density functions based on Bayesian nonparametric priors. Additionally, the functional form of the system is estimated by leveraging Bayesian neural networks which also leads to flexible uncertainty quantification. Asymptotically on the number of hidden neurons, the proposed model converges to full nonparametric Bayesian regression model. A Gibbs sampler for posterior inference is proposed and its effectiveness is illustrated on simulated and real time series.

stat.ME↗

arXiv2022

Temporal Parallelisation of Dynamic Programming and Linear Quadratic Control

This paper proposes a general formulation for temporal parallelisation of dynamic programming for optimal control problems. We derive the elements and associative operators to be able to use parallel scans to solve these problems with logarithmic time complexity rather than linear time complexity. We apply this methodology to problems with finite state and control spaces, linear quadratic tracking control problems, and to a class of nonlinear control problems. The computational benefits of the parallel methods are demonstrated via numerical simulations run on a graphics processing unit.

math.OC↗

arXiv2021

Non-linear Gaussian smoothing with Taylor moment expansion

This letter is concerned with solving continuous-discrete Gaussian smoothing problems by using the Taylor moment expansion (TME) scheme. In the proposed smoothing method, we apply the TME method to approximate the transition density of the stochastic differential equation in the dynamic model. Furthermore, we derive a theoretical error bound (in the mean square sense) of the TME smoothing estimates showing that the smoother is stable under weak assumptions. Numerical experiments are presented in order to illustrate practical use of the method.

math.NA↗