Jason L. Williams

Jason L. Williams
Are you Jason L. Williams?

Claim your profile, edit publications, add additional information:

Contact Details

Name
Jason L. Williams
Affiliation
Location

Pubs By Year

Pub Categories

 
Computer Science - Artificial Intelligence (5)
 
Computer Science - Computer Vision and Pattern Recognition (4)
 
Physics - Atomic Physics (3)
 
Physics - Space Physics (2)
 
Quantum Physics (2)
 
Computer Science - Computation and Language (2)
 
General Relativity and Quantum Cosmology (2)
 
Computer Science - Learning (2)
 
Mathematics - Optimization and Control (1)
 
Mathematics - Information Theory (1)
 
Computer Science - Information Theory (1)
 
Statistics - Methodology (1)
 
Statistics - Machine Learning (1)

Publications Authored By Jason L. Williams

We provide a derivation of the Poisson multi-Bernoulli mixture (PMBM) filter for multi-target tracking with the standard point target measurements without using probability generating functionals or functional derivatives. We also establish the connection with the \delta-generalised labelled multi-Bernoulli (\delta-GLMB) filter, showing that a \delta-GLMB density represents a multi-Bernoulli mixture with labelled targets so it can be seen as a special case of PMBM. In addition, we propose an implementation for linear/Gaussian dynamic and measurement models and how to efficiently obtain typical estimators in the literature from the PMBM. Read More

End-to-end learning of recurrent neural networks (RNNs) is an attractive solution for dialog systems; however, current techniques are data-intensive and require thousands of dialogs to learn simple behaviors. We introduce Hybrid Code Networks (HCNs), which combine an RNN with domain-specific knowledge encoded as software and system action templates. Compared to existing end-to-end approaches, HCNs considerably reduce the amount of training data required, while retaining the key benefit of inferring a latent representation of dialog state. Read More

Representing a dialog policy as a recurrent neural network (RNN) is attractive because it handles partial observability, infers a latent representation of state, and can be optimized with supervised learning (SL) or reinforcement learning (RL). For RL, a policy gradient approach is natural, but is sample inefficient. In this paper, we present 3 methods for reducing the number of dialogs required to optimize an RNN-based dialog policy with RL. Read More

The gravity gradient is one of the most serious systematic effects in atomic tests of the equivalence principle (EP). While differential acceleration measurements performed with different atomic species under free fall test the validity of EP, minute displacements between the test masses in a gravity gradient produces a false EP-violating signal that limits the precision of the test. We show that gravity inversion and modulation using a gimbal mount can suppress the systematics due to gravity gradients caused by both moving and stationary parts of the instrument as well as the environment, strongly reducing the need to overlap two species. Read More

Data association, the reasoning over correspondence between targets and measurements, is a problem of fundamental importance in target tracking. Recently, belief propagation (BP) has emerged as a promising method for estimating the marginal probabilities of target/measurement association, providing fast, accurate estimates. The excellent performance of BP in the particular formulation used may be attributed to the convexity of the underlying free energy which it implicitly optimises. Read More

This paper presents a model for end-to-end learning of task-oriented dialog systems. The main component of the model is a recurrent neural network (an LSTM), which maps from raw dialog history directly to a distribution over system actions. The LSTM automatically infers a representation of dialog history, which relieves the system developer of much of the manual feature engineering of dialog state. Read More

We describe the Quantum Test of the Equivalence principle and Space Time (QTEST), a concept for an atom interferometry mission on the International Space Station (ISS). The primary science objective of the mission is a test of Einstein's equivalence principle with two rubidium isotope gases at a precision of better than 10$^{-15}$. Distinct from the classical tests is the use of quantum wave packets and their expected large spatial separation in the QTEST experiment. Read More

High sensitivity differential atom interferometers are promising for precision measurements in science frontiers in space, including gravity field mapping for Earth science studies and gravitational wave detection. We propose a new configuration of twin atom interferometers connected by a laser ranging interferometer (LRI-AI) to provide precise information of the displacements between the two AI reference mirrors and a means to phase-lock the two independent interferometer lasers over long distances, thereby further enhancing the feasibility of long baseline differential atom interferometers. We show that a properly implemented LRI-AI can achieve equivalent functionality to the conventional differential atom interferometer measurement system. Read More

The joint probabilistic data association (JPDA) filter is a popular tracking methodology for problems involving well-spaced targets, but it is rarely applied in problems with closely-spaced targets due to its complexity in these cases, and due to the well-known phenomenon of coalescence. This paper addresses these difficulties using random finite sets (RFSs) and variational inference, deriving a highly tractable, approximate method for obtaining the multi-Bernoulli distribution that minimises the set Kullback-Leibler (KL) divergence from the true posterior, working within the RFS framework to incorporate uncertainty in target existence. The derivation is interpreted as an application of expectation-maximisation (EM), where the missing data is the correspondence of Bernoulli components (i. Read More

Data association, the problem of reasoning over correspondence between targets and measurements, is a fundamental problem in tracking. This paper presents a graphical model formulation of data association and applies an approximate inference method, belief propagation (BP), to obtain estimates of marginal association probabilities. We prove that BP is guaranteed to converge, and bound the number of iterations necessary. Read More

The probability hypothesis density (PHD) and multi-target multi-Bernoulli (MeMBer) filters are two leading algorithms that have emerged from random finite sets (RFS). In this paper we study a method which combines these two approaches. Our work is motivated by a sister paper, which proves that the full Bayes RFS filter naturally incorporates a Poisson component representing targets that have never been detected, and a linear combination of multi-Bernoulli components representing targets under track. Read More

Recent developments in random finite sets (RFSs) have yielded a variety of tracking methods that avoid data association. This paper derives a form of the full Bayes RFS filter and observes that data association is implicitly present, in a data structure similar to MHT. Subsequently, algorithms are obtained by approximating the distribution of associations. Read More

Random finite sets (RFSs) has been a fruitful area of research in recent years, yielding new approximate filters such as the probability hypothesis density (PHD), cardinalised PHD (CPHD), and multiple target multi-Bernoulli (MeMBer). These new methods have largely been based on approximations that side-step the need for measurement-to-track association. Comparably, RFS methods that incorporate data association, such as Morelande and Challa's (M-C) method, have received little attention. Read More