arXiv subjects

Simon Maskell

Publications and source records attributed to Simon Maskell.

At least 37 records · Page 2Linked to original sources

arXiv2023

Parallel Approaches to Accelerate Bayesian Decision Trees

Markov Chain Monte Carlo (MCMC) is a well-established family of algorithms primarily used in Bayesian statistics to sample from a target distribution when direct sampling is challenging. Existing work on Bayesian decision trees uses MCMC. Unfortunately, this can be slow, especially when considering large volumes of data. It is hard to parallelise the accept-reject component of the MCMC. None-the-less, we propose two methods for exploiting parallelism in the MCMC: in the first, we replace the MCMC with another numerical Bayesian approach, the Sequential Monte Carlo (SMC) sampler, which has the appealing property that it is an inherently parallel algorithm; in the second, we consider data partitioning. Both methods use multi-core processing with a HighPerformance Computing (HPC) resource. We test the two methods in various study settings to determine which method is the most beneficial for each test case. Experiments show that data partitioning has limited utility in the settings we consider and that the use of the SMC sampler can improve run-time (compared to the sequential implementation) by up to a factor of 343.

stat.CO↗

arXiv2022

Data-driven clustering and Bernoulli merging for the Poisson multi-Bernoulli mixture filter

This paper proposes a clustering and merging approach for the Poisson multi-Bernoulli mixture (PMBM) filter to lower its computational complexity and make it suitable for multiple target tracking with a high number of targets. We define a measurement-driven clustering algorithm to reduce the data association problem into several subproblems, and we provide the derivation of the resulting clustered PMBM posterior density via Kullback-Leibler divergence minimisation. Furthermore, we investigate different strategies to reduce the number of single target hypotheses by approximating the posterior via merging and inter-track swapping of Bernoulli components. We evaluate the performance of the proposed algorithm on simulated tracking scenarios with more than one thousand targets.

eess.SP↗

arXiv2022

Can We Automate the Analysis of Online Child Sexual Exploitation Discourse?

Social media's growing popularity raises concerns around children's online safety. Interactions between minors and adults with predatory intentions is a particularly grave concern. Research into online sexual grooming has often relied on domain experts to manually annotate conversations, limiting both scale and scope. In this work, we test how well-automated methods can detect conversational behaviors and replace an expert human annotator. Informed by psychological theories of online grooming, we label $6772$ chat messages sent by child-sex offenders with one of eleven predatory behaviors. We train bag-of-words and natural language inference models to classify each behavior, and show that the best performing models classify behaviors in a manner that is consistent, but not on-par, with human annotation.

cs.CL↗

arXiv2022

Inference of Stochastic Disease Transmission Models Using Particle-MCMC and a Gradient Based Proposal

State-space models have been widely used to model the dynamics of communicable diseases in populations of interest by fitting to time-series data. Particle filters have enabled these models to incorporate stochasticity and so can better reflect the true nature of population behaviours. Relevant parameters such as the spread of the disease, $R_t$, and recovery rates can be inferred using Particle MCMC. The standard method uses a Metropolis-Hastings random-walk proposal which can struggle to reach the stationary distribution in a reasonable time when there are multiple parameters. In this paper we obtain full Bayesian parameter estimations using gradient information and the No U-Turn Sampler (NUTS) when proposing new parameters of stochastic non-linear Susceptible-Exposed-Infected-Recovered (SEIR) and SIR models. Although NUTS makes more than one target evaluation per iteration, we show that it can provide more accurate estimates in a shorter run time than Metropolis-Hastings.

stat.AP↗

arXiv2022

Efficient Learning of the Parameters of Non-Linear Models using Differentiable Resampling in Particle Filters

It has been widely documented that the sampling and resampling steps in particle filters cannot be differentiated. The {\itshape reparameterisation trick} was introduced to allow the sampling step to be reformulated into a differentiable function. We extend the {\itshape reparameterisation trick} to include the stochastic input to resampling therefore limiting the discontinuities in the gradient calculation after this step. Knowing the gradients of the prior and likelihood allows us to run particle Markov Chain Monte Carlo (p-MCMC) and use the No-U-Turn Sampler (NUTS) as the proposal when estimating parameters. We compare the Metropolis-adjusted Langevin algorithm (MALA), Hamiltonian Monte Carlo with different number of steps and NUTS. We consider two state-space models and show that NUTS improves the mixing of the Markov chain and can produce more accurate results in less computational time.

stat.ML↗

arXiv2022

Position fixing with cold atom gravity gradiometers

This paper proposes a position fixing method for autonomous navigation using partial gravity gradient solutions from cold atom interferometers. Cold atom quantum sensors can provide ultra-precise measurements of inertial quantities, such as acceleration and rotation rates. However, we investigate the use of pairs of cold atom interferometers to measure the local gravity gradient and to provide position information by referencing these measurements against a suitable database. Simulating the motion of a vehicle, we use partial gravity gradient measurements to reduce the positional drift associated with inertial navigation systems. Using standard open source global gravity databases, we show stable navigation solutions for trajectories of over 1000km.

quant-ph↗

arXiv2022

Classical Tracking for Quantum Trajectories

Quantum state estimation, based on the numerical integration of stochastic master equations (SMEs), provides estimates for the evolution of quantum systems subject to continuous weak measurements. The approach is similar to classical state estimation methods in that the quantum trajectories produced by solving the SME are conditioned on continuous classical measurement signals. In this paper, we explore the use of classical state estimation for a candidate quantum system, one based on an experimentally realisable system: a material object undergoing continuous feedback cooling in an optical trap. In particular, we demonstrate that classical tracking methods based on particle filters can be used to track quantum states, and are particularly useful for higher temperature regimes where quantum state estimation becomes computationally demanding.

quant-ph↗

arXiv2022

Ensemble Kalman filter based Sequential Monte Carlo Sampler for sequential Bayesian inference

Many real-world problems require one to estimate parameters of interest, in a Bayesian framework, from data that are collected sequentially in time. Conventional methods for sampling from posterior distributions, such as {Markov Chain Monte Carlo} can not efficiently address such problems as they do not take advantage of the data's sequential structure. To this end, sequential methods which seek to update the posterior distribution whenever a new collection of data become available are often used to solve these types of problems. Two popular choices of sequential method are the Ensemble Kalman filter (EnKF) and the sequential Monte Carlo sampler (SMCS). While EnKF only computes a Gaussian approximation of the posterior distribution, SMCS can draw samples directly from the posterior. Its performance, however, depends critically upon the kernels that are used. In this work, we present a method that constructs the kernels of SMCS using an EnKF formulation, and we demonstrate the performance of the method with numerical examples.

stat.ME↗

arXiv2021

Fusing Low-Latency Data Feeds with Death Data to Accurately Nowcast COVID-19 Related Deaths

The emergence of the novel coronavirus (COVID-19) has generated a need to quickly and accurately assemble up-to-date information related to its spread. While it is possible to use deaths to provide a reliable information feed, the latency of data derived from deaths is significant. Confirmed cases derived from positive test results potentially provide a lower latency data feed. However, the sampling of those tested varies with time and the reason for testing is often not recorded. Hospital admissions typically occur around 1-2 weeks after infection and can be considered out of date in relation to the time of initial infection. The extent to which these issues are problematic is likely to vary over time and between countries. We use a machine learning algorithm for natural language processing, trained in multiple languages, to identify symptomatic individuals derived from social media and, in particular Twitter, in real-time. We then use an extended SEIRD epidemiological model to fuse combinations of low-latency feeds, including the symptomatic counts from Twitter, with death data to estimate parameters of the model and nowcast the number of people in each compartment. The model is implemented in the probabilistic programming language Stan and uses a bespoke numerical integrator. We present results showing that using specific low-latency data feeds along with death data provides more consistent and accurate forecasts of COVID-19 related deaths than using death data alone.

stat.AP↗

arXiv2021

An analysis on metric-driven multi-target sensor management: GOSPA versus OSPA

This paper presents an analysis on sensor management using a cost function based on a multi-target metric, in particular, the optimal subpattern-assignment (OSPA) metric, the unnormalised OSPA (UOSPA) metric and the generalised OSPA (GOSPA) metric (α=2). We consider the problem of managing an array of sensors, where each sensor is able to observe a region of the surveillance area, not covered by other sensors, with a given sensing cost. We look at the case in which there are far-away, independent potential targets, at maximum one per sensor region. In this set-up, the optimal action using GOSPA is taken for each sensor independently, as we may expect. On the contrary, as a consequence of the spooky effect at a distance in optimal OSPA/UOSPA estimation, the optimal actions for different sensors using OSPA and UOSPA are entangled.

eess.SY↗

arXiv2021

The No-U-Turn Sampler as a Proposal Distribution in a Sequential Monte Carlo Sampler with a Near-Optimal L-Kernel

Markov Chain Monte Carlo (MCMC) is a powerful method for drawing samples from non-standard probability distributions and is utilized across many fields and disciplines. Methods such as Metropolis-Adjusted Langevin (MALA) and Hamiltonian Monte Carlo (HMC), which use gradient information to explore the target distribution, are popular variants of MCMC. The Sequential Monte Carlo (SMC) sampler is an alternative sampling method which, unlike MCMC, can readily utilise parallel computing architectures and also has tuning parameters not available to MCMC. One such parameter is the L-kernel which can be used to minimise the variance of the estimates from an SMC sampler. In this letter, we show how the proposal used in the No-U-Turn Sampler (NUTS), an advanced variant of HMC, can be incorporated into an SMC sampler to improve the efficiency of the exploration of the target space. We also show how the SMC sampler can be optimized using both a near-optimal L-kernel and a Hamiltonian proposal

stat.CO↗

arXiv2020

Increasing the efficiency of Sequential Monte Carlo samplers through the use of approximately optimal L-kernels

By facilitating the generation of samples from arbitrary probability distributions, Markov Chain Monte Carlo (MCMC) is, arguably, \emph{the} tool for the evaluation of Bayesian inference problems that yield non-standard posterior distributions. In recent years, however, it has become apparent that Sequential Monte Carlo (SMC) samplers have the potential to outperform MCMC in a number of ways. SMC samplers are better suited to highly parallel computing architectures and also feature various tuning parameters that are not available to MCMC. One such parameter - the `L-kernel' - is a user-defined probability distribution that can be used to influence the efficiency of the sampler. In the current paper, the authors explain how to derive an expression for the L-kernel that minimises the variance of the estimates realised by an SMC sampler. Various approximation methods are then proposed to aid implementation of the proposed L-kernel. The improved performance of the resulting algorithm is demonstrated in multiple scenarios. For the examples shown in the current paper, the use of an approximately optimum L-kernel has reduced the variance of the SMC estimates by up to 99 % while also reducing the number of times that resampling was required by between 65 % and 70 %. Python code and code tests accompanying this manuscript are available through the Github repository \url{https://github.com/plgreenLIRU/SMC_approx_optL}.

stat.CO↗

arXiv2020

Reliability Validation of Learning Enabled Vehicle Tracking

This paper studies the reliability of a real-world learning-enabled system, which conducts dynamic vehicle tracking based on a high-resolution wide-area motion imagery input. The system consists of multiple neural network components -- to process the imagery inputs -- and multiple symbolic (Kalman filter) components -- to analyse the processed information for vehicle tracking. It is known that neural networks suffer from adversarial examples, which make them lack robustness. However, it is unclear if and how the adversarial examples over learning components can affect the overall system-level reliability. By integrating a coverage-guided neural network testing tool, DeepConcolic, with the vehicle tracking system, we found that (1) the overall system can be resilient to some adversarial examples thanks to the existence of other components, and (2) the overall system presents an extra level of uncertainty which cannot be determined by analysing the deep learning components only. This research suggests the need for novel verification and validation methods for learning-enabled systems.

cs.CV↗

arXiv2019

Detecting and Tracking Small Moving Objects in Wide Area Motion Imagery (WAMI) Using Convolutional Neural Networks (CNNs)

This paper proposes an approach to detect moving objects in Wide Area Motion Imagery (WAMI), in which the objects are both small and well separated. Identifying the objects only using foreground appearance is difficult since a $100-$pixel vehicle is hard to distinguish from objects comprising the background. Our approach is based on background subtraction as an efficient and unsupervised method that is able to output the shape of objects. In order to reliably detect low contrast and small objects, we configure the background subtraction to extract foreground regions that might be objects of interest. While this dramatically increases the number of false alarms, a Convolutional Neural Network (CNN) considering both spatial and temporal information is then trained to reject the false alarms. In areas with heavy traffic, the background subtraction yields merged detections. To reduce the complexity of multi-target tracker needed, we train another CNN to predict the positions of multiple moving objects in an area. Our approach shows competitive detection performance on smaller objects relative to the state-of-the-art. We adopt a GM-PHD filter to associate detections over time and analyse the resulting performance.

cs.CV↗

arXiv2019

Robust and Accurate Global Motion Estimation Using the Student-t Distribution

Pixel-based Global Motion Estimation (GME) has always struggled to simultaneously reject outliers, avoid local minima and run quickly. There are many robust cost functions that perform well in terms of rejecting outliers, but they can yield unstable results during long image sequences as a result of their inability to adjust to changes in image content. In this letter, we propose a parameterised student-t cost function that can interpolate between two cost functions that are amongst the most widely in image registration problems, the L2 norm and the Cauchy-Lorentzian function. We also propose a parameter estimation method that helps to find the best parameters for the proposed cost function. Experiments prove that the proposed approach can estimate global motion accurately relative to the existing cost functions without demanding a higher computational cost.

eess.IV↗

arXiv2019

A Single SMC Sampler on MPI that Outperforms a Single MCMC Sampler

Markov Chain Monte Carlo (MCMC) is a well-established family of algorithms which are primarily used in Bayesian statistics to sample from a target distribution when direct sampling is challenging. Single instances of MCMC methods are widely considered hard to parallelise in a problem-agnostic fashion and hence, unsuitable to meet both constraints of high accuracy and high throughput. Sequential Monte Carlo (SMC) Samplers can address the same problem, but are parallelisable: they share with Particle Filters the same key tasks and bottleneck. Although a rich literature already exists on MCMC methods, SMC Samplers are relatively underexplored, such that no parallel implementation is currently available. In this paper, we first propose a parallel MPI version of the SMC Sampler, including an optimised implementation of the bottleneck, and then compare it with single-core Metropolis-Hastings. The goal is to show that SMC Samplers may be a promising alternative to MCMC methods with high potential for future improvements. We demonstrate that a basic SMC Sampler with 512 cores is up to 85 times faster or up to 8 times more accurate than Metropolis-Hastings.

stat.CO↗

arXiv2018

Convolutional Neural Networks for Aerial Vehicle Detection and Recognition

This paper investigates the problem of aerial vehicle recognition using a text-guided deep convolutional neural network classifier. The network receives an aerial image and a desired class, and makes a yes or no output by matching the image and the textual description of the desired class. We train and test our model on a synthetic aerial dataset and our desired classes consist of the combination of the class types and colors of the vehicles. This strategy helps when considering more classes in testing than in training.

cs.CV↗

arXiv2018

Dynamical model selection near the quantum-classical boundary

We discuss a general method of model selection from experimentally recorded time-trace data. This method can be used to distinguish between quantum and classical dynamical models. It can be used in post-selection as well as for real-time analysis, and offers an alternative to statistical tests based on state-reconstruction methods. We examine the conditions that optimize quantum hypothesis testing, maximizing one's ability to discriminate between classical and quantum models. We set upper limits on the temperature and lower limits on the measurement efficiencies required to explore these differences, using a novel experiment in levitated optomechanical systems as an example.

quant-ph↗