Showing 1–50 of 57 results for author: Kuipers, J

Exact discovery is polynomial for sparse causal Bayesian networks

Authors: Felix L. Rios, Giusi Moffa, Jack Kuipers

Abstract: Causal Bayesian networks are widely used tools for summarising the dependencies between variables and elucidating their putative causal relationships. Learning networks from data is computationally hard in general. The current state-of-the-art approaches for exact causal discovery are integer linear programming over the underlying space of directed acyclic graphs, dynamic programming and shortest-… ▽ More Causal Bayesian networks are widely used tools for summarising the dependencies between variables and elucidating their putative causal relationships. Learning networks from data is computationally hard in general. The current state-of-the-art approaches for exact causal discovery are integer linear programming over the underlying space of directed acyclic graphs, dynamic programming and shortest-path searches over the space of topological orders, and constraint programming combining both. For dynamic programming over orders, the computational complexity is known to be exponential base 2 in the number of variables in the network. We demonstrate how to use properties of Bayesian networks to prune the search space and lower the computational cost, while still guaranteeing exact discovery. When including new path-search and divide-and-conquer criteria, we prove optimality in quadratic time for matchings, and polynomial time for any network class with logarithmically-bound largest connected components. In simulation studies we observe the polynomial dependence for sparse networks and that, beyond some critical value, the logarithm of the base grows with the network density. Our approach then out-competes the state-of-the-art at lower densities. These results therefore pave the way for faster exact causal discovery in larger and sparser networks. △ Less

Submitted 21 June, 2024; originally announced June 2024.

Comments: 14 pages, 3 figures with 8 pages of Supplement. Code available at https://github.com/felixleopoldo/dncDagger

A new way to evaluate G-Wishart normalising constants via Fourier analysis

Authors: Ching Wong, Giusi Moffa, Jack Kuipers

Abstract: The G-Wishart distribution is an essential component for the Bayesian analysis of Gaussian graphical models as the conjugate prior for the precision matrix. Evaluating the marginal likelihood of such models usually requires computing high-dimensional integrals to determine the G-Wishart normalising constant. Closed-form results are known for decomposable or chordal graphs, while an explicit repres… ▽ More The G-Wishart distribution is an essential component for the Bayesian analysis of Gaussian graphical models as the conjugate prior for the precision matrix. Evaluating the marginal likelihood of such models usually requires computing high-dimensional integrals to determine the G-Wishart normalising constant. Closed-form results are known for decomposable or chordal graphs, while an explicit representation as a formal series expansion has been derived recently for general graphs. The nested infinite sums, however, do not lend themselves to computation, remaining of limited practical value. Borrowing techniques from random matrix theory and Fourier analysis, we provide novel exact results well suited to the numerical evaluation of the normalising constant for a large class of graphs beyond chordal graphs. Furthermore, they open new possibilities for develo** more efficient sampling schemes for Bayesian inference of Gaussian graphical models. △ Less

Submitted 10 April, 2024; originally announced April 2024.

MSC Class: 62H05; 60E05 (Primary) 62F15 (Secondary)

Bayesian Causal Inference with Gaussian Process Networks

Authors: Enrico Giudice, Jack Kuipers, Giusi Moffa

Abstract: Causal discovery and inference from observational data is an essential problem in statistics posing both modeling and computational challenges. These are typically addressed by imposing strict assumptions on the joint distribution such as linearity. We consider the problem of the Bayesian estimation of the effects of hypothetical interventions in the Gaussian Process Network (GPN) model, a flexibl… ▽ More Causal discovery and inference from observational data is an essential problem in statistics posing both modeling and computational challenges. These are typically addressed by imposing strict assumptions on the joint distribution such as linearity. We consider the problem of the Bayesian estimation of the effects of hypothetical interventions in the Gaussian Process Network (GPN) model, a flexible causal framework which allows describing the causal relationships nonparametrically. We detail how to perform causal inference on GPNs by simulating the effect of an intervention across the whole network and propagating the effect of the intervention on downstream variables. We further derive a simpler computational approximation by estimating the intervention distribution as a function of local variables only, modeling the conditional distributions via additive Gaussian processes. We extend both frameworks beyond the case of a known causal graph, incorporating uncertainty about the causal structure via Markov chain Monte Carlo methods. Simulation studies show that our approach is able to identify the effects of hypothetical interventions with non-Gaussian, non-linear observational data and accurately reflect the posterior uncertainty of the causal estimates. Finally we compare the results of our GPN-based causal inference approach to existing methods on a dataset of $A.~thaliana$ gene expressions. △ Less

Submitted 1 February, 2024; originally announced February 2024.

Fair Clustering: A Causal Perspective

Authors: Fritz Bayer, Drago Plecko, Niko Beerenwinkel, Jack Kuipers

Abstract: Clustering algorithms may unintentionally propagate or intensify existing disparities, leading to unfair representations or biased decision-making. Current fair clustering methods rely on notions of fairness that do not capture any information on the underlying causal mechanisms. We show that optimising for non-causal fairness notions can paradoxically induce direct discriminatory effects from a c… ▽ More Clustering algorithms may unintentionally propagate or intensify existing disparities, leading to unfair representations or biased decision-making. Current fair clustering methods rely on notions of fairness that do not capture any information on the underlying causal mechanisms. We show that optimising for non-causal fairness notions can paradoxically induce direct discriminatory effects from a causal standpoint. We present a clustering approach that incorporates causal fairness metrics to provide a more nuanced approach to fairness in unsupervised learning. Our approach enables the specification of the causal fairness metrics that should be minimised. We demonstrate the efficacy of our methodology using datasets known to harbour unfair biases. △ Less

Submitted 14 December, 2023; originally announced December 2023.

Comments: 14 pages

A Bayesian Take on Gaussian Process Networks

Authors: Enrico Giudice, Jack Kuipers, Giusi Moffa

Abstract: Gaussian Process Networks (GPNs) are a class of directed graphical models which employ Gaussian processes as priors for the conditional expectation of each variable given its parents in the network. The model allows the description of continuous joint distributions in a compact but flexible manner with minimal parametric assumptions on the dependencies between variables. Bayesian structure learnin… ▽ More Gaussian Process Networks (GPNs) are a class of directed graphical models which employ Gaussian processes as priors for the conditional expectation of each variable given its parents in the network. The model allows the description of continuous joint distributions in a compact but flexible manner with minimal parametric assumptions on the dependencies between variables. Bayesian structure learning of GPNs requires computing the posterior over graphs of the network and is computationally infeasible even in low dimensions. This work implements Monte Carlo and Markov Chain Monte Carlo methods to sample from the posterior distribution of network structures. As such, the approach follows the Bayesian paradigm, comparing models via their marginal likelihood and computing the posterior probability of the GPN features. Simulation studies show that our method outperforms state-of-the-art algorithms in recovering the graphical structure of the network and provides an accurate approximation of its posterior distribution. △ Less

Submitted 24 November, 2023; v1 submitted 20 June, 2023; originally announced June 2023.

Effect of a tunnel barrier on time delay statistics

Authors: Marcel Novaes, Jack Kuipers

Abstract: We develop a semiclassical approach for the statistics of the time delay in quantum chaotic systems in the presence of a tunnel barrier, for broken time-reversal symmetry. Results are obtained as asymptotic series in powers of the reflectivity of the barrier, with coefficients that are rational functions of the channel number. Exact expressions, valid for arbitrary reflectivity and channel number,… ▽ More We develop a semiclassical approach for the statistics of the time delay in quantum chaotic systems in the presence of a tunnel barrier, for broken time-reversal symmetry. Results are obtained as asymptotic series in powers of the reflectivity of the barrier, with coefficients that are rational functions of the channel number. Exact expressions, valid for arbitrary reflectivity and channel number, are conjectured and numerically verified for specific families of statistical moments. △ Less

Submitted 17 February, 2023; originally announced February 2023.

Comments: 9 pages, 2 figures

The interventional Bayesian Gaussian equivalent score for Bayesian causal inference with unknown soft interventions

Authors: Jack Kuipers, Giusi Moffa

Abstract: Describing the causal relations governing a system is a fundamental task in many scientific fields, ideally addressed by experimental studies. However, obtaining data under intervention scenarios may not always be feasible, while discovering causal relations from purely observational data is notoriously challenging. In certain settings, such as genomics, we may have data from heterogeneous study c… ▽ More Describing the causal relations governing a system is a fundamental task in many scientific fields, ideally addressed by experimental studies. However, obtaining data under intervention scenarios may not always be feasible, while discovering causal relations from purely observational data is notoriously challenging. In certain settings, such as genomics, we may have data from heterogeneous study conditions, with soft (partial) interventions only pertaining to a subset of the study variables, whose effects and targets are possibly unknown. Combining data from experimental and observational studies offers the opportunity to leverage both domains and improve on the identifiability of causal structures. To this end, we define the interventional BGe score for a mixture of observational and interventional data, where the targets and effects of intervention may be unknown. To demonstrate the approach we compare its performance to other state-of-the-art algorithms, both in simulations and data analysis applications. Prerogative of our method is that it takes a Bayesian perspective leading to a full characterisation of the posterior distribution of the DAG structures. Given a sample of DAGs one can also automatically derive full posterior distributions of the intervention effects. Consequently the method effectively captures the uncertainty both in the structure and the parameter estimates. Codes to reproduce the simulations and analyses are publicly available at github.com/jackkuipers/iBGe △ Less

Submitted 5 May, 2022; originally announced May 2022.

High-Dimensional Inference in Bayesian Networks

Authors: Fritz M. Bayer, Giusi Moffa, Niko Beerenwinkel, Jack Kuipers

Abstract: Inference of the marginal probability distribution is defined as the calculation of the probability of a subset of the variables and is relevant for handling missing data and hidden variables. While inference of the marginal probability distribution is crucial for various problems in machine learning and statistics, its exact computation is generally not feasible for categorical variables in Bayes… ▽ More Inference of the marginal probability distribution is defined as the calculation of the probability of a subset of the variables and is relevant for handling missing data and hidden variables. While inference of the marginal probability distribution is crucial for various problems in machine learning and statistics, its exact computation is generally not feasible for categorical variables in Bayesian networks due to the NP-hardness of this task. We develop a divide-and-conquer approach using the graphical properties of Bayesian networks to split the computation of the marginal probability distribution into sub-calculations of lower dimensionality, thus reducing the overall computational complexity. Exploiting this property, we present an efficient and scalable algorithm for calculating the marginal probability distribution for categorical variables. The novel method is compared against state-of-the-art approximate inference methods in a benchmarking study, where it displays superior performance. As an immediate application, we demonstrate how our method can be used to classify incomplete data against Bayesian networks and use this approach for identifying the cancer subtype of kidney cancer patient samples. △ Less

Submitted 20 July, 2022; v1 submitted 16 December, 2021; originally announced December 2021.

The Dual PC Algorithm and the Role of Gaussianity for Structure Learning of Bayesian Networks

Authors: Enrico Giudice, Jack Kuipers, Giusi Moffa

Abstract: Learning the graphical structure of Bayesian networks is key to describing data-generating mechanisms in many complex applications but poses considerable computational challenges. Observational data can only identify the equivalence class of the directed acyclic graph underlying a Bayesian network model, and a variety of methods exist to tackle the problem. Under certain assumptions, the popular P… ▽ More Learning the graphical structure of Bayesian networks is key to describing data-generating mechanisms in many complex applications but poses considerable computational challenges. Observational data can only identify the equivalence class of the directed acyclic graph underlying a Bayesian network model, and a variety of methods exist to tackle the problem. Under certain assumptions, the popular PC algorithm can consistently recover the correct equivalence class by reverse-engineering the conditional independence (CI) relationships holding in the variable distribution. The dual PC algorithm is a novel scheme to carry out the CI tests within the PC algorithm by leveraging the inverse relationship between covariance and precision matrices. By exploiting block matrix inversions we can also perform tests on partial correlations of complementary (or dual) conditioning sets. The multiple CI tests of the dual PC algorithm proceed by first considering marginal and full-order CI relationships and progressively moving to central-order ones. Simulation studies show that the dual PC algorithm outperforms the classic PC algorithm both in terms of run time and in recovering the underlying network structure, even in the presence of deviations from Gaussianity. Additionally, we show that the dual PC algorithm applies for Gaussian copula models, and demonstrate its performance in that setting. △ Less

Submitted 27 June, 2023; v1 submitted 16 December, 2021; originally announced December 2021.

Benchpress: A Scalable and Versatile Workflow for Benchmarking Structure Learning Algorithms

Authors: Felix L. Rios, Giusi Moffa, Jack Kuipers

Abstract: Describing the relationship between the variables in a study domain and modelling the data generating mechanism is a fundamental problem in many empirical sciences. Probabilistic graphical models are one common approach to tackle the problem. Learning the graphical structure for such models is computationally challenging and a fervent area of current research with a plethora of algorithms being de… ▽ More Describing the relationship between the variables in a study domain and modelling the data generating mechanism is a fundamental problem in many empirical sciences. Probabilistic graphical models are one common approach to tackle the problem. Learning the graphical structure for such models is computationally challenging and a fervent area of current research with a plethora of algorithms being developed. To facilitate the benchmarking of different methods, we present a novel Snakemake workflow, called Benchpress for producing scalable, reproducible, and platform-independent benchmarks of structure learning algorithms for probabilistic graphical models. Benchpress is interfaced via a simple JSON-file, which makes it accessible for all users, while the code is designed in a fully modular fashion to enable researchers to contribute additional methodologies. Benchpress currently provides an interface to a large number of state-of-the-art algorithms from libraries such as BDgraph, BiDAG, bnlearn, causal-learn, gCastle, GOBNILP, pcalg, r.blip, scikit-learn, TETRAD, and trilearn as well as a variety of methods for data generating models and performance evaluation. Alongside user-defined models and randomly generated datasets, the workflow also includes a number of standard datasets and graphical models from the literature, which may be included in a benchmarking study. We demonstrate the applicability of this workflow for learning Bayesian networks in five typical data scenarios. The source code and documentation is publicly available from http://benchpressdocs.readthedocs.io. △ Less

Submitted 4 December, 2023; v1 submitted 8 July, 2021; originally announced July 2021.

Comments: 41 pages, 8 figure

Bayesian structure learning and sampling of Bayesian networks with the R package BiDAG

Authors: Polina Suter, Jack Kuipers, Giusi Moffa, Niko Beerenwinkel

Abstract: The R package BiDAG implements Markov chain Monte Carlo (MCMC) methods for structure learning and sampling of Bayesian networks. The package includes tools to search for a maximum a posteriori (MAP) graph and to sample graphs from the posterior distribution given the data. A new hybrid approach to structure learning enables inference in large graphs. In the first step, we define a reduced search s… ▽ More The R package BiDAG implements Markov chain Monte Carlo (MCMC) methods for structure learning and sampling of Bayesian networks. The package includes tools to search for a maximum a posteriori (MAP) graph and to sample graphs from the posterior distribution given the data. A new hybrid approach to structure learning enables inference in large graphs. In the first step, we define a reduced search space by means of the PC algorithm or based on prior knowledge. In the second step, an iterative order MCMC scheme proceeds to optimize within the restricted search space and estimate the MAP graph. Sampling from the posterior distribution is implemented using either order or partition MCMC. The models and algorithms can handle both discrete and continuous data. The BiDAG package also provides an implementation of MCMC schemes for structure learning and sampling of dynamic Bayesian networks. △ Less

Submitted 2 May, 2021; originally announced May 2021.

Learning Bayesian Networks from Ordinal Data

Authors: Xiang Ge Luo, Giusi Moffa, Jack Kuipers

Abstract: Bayesian networks are a powerful framework for studying the dependency structure of variables in a complex system. The problem of learning Bayesian networks is tightly associated with the given data type. Ordinal data, such as stages of cancer, rating scale survey questions, and letter grades for exams, are ubiquitous in applied research. However, existing solutions are mainly for continuous and n… ▽ More Bayesian networks are a powerful framework for studying the dependency structure of variables in a complex system. The problem of learning Bayesian networks is tightly associated with the given data type. Ordinal data, such as stages of cancer, rating scale survey questions, and letter grades for exams, are ubiquitous in applied research. However, existing solutions are mainly for continuous and nominal data. In this work, we propose an iterative score-and-search method - called the Ordinal Structural EM (OSEM) algorithm - for learning Bayesian networks from ordinal data. Unlike traditional approaches designed for nominal data, we explicitly respect the ordering amongst the categories. More precisely, we assume that the ordinal variables originate from marginally discretizing a set of Gaussian variables, whose structural dependence in the latent space follows a directed acyclic graph. Then, we adopt the Structural EM algorithm and derive closed-form scoring functions for efficient graph searching. Through simulation studies, we illustrate the superior performance of the OSEM algorithm compared to the alternatives and analyze various factors that may influence the learning accuracy. Finally, we demonstrate the practicality of our method with a real-world application on psychological survey data from 408 patients with co-morbid symptoms of obsessive-compulsive disorder and depression. △ Less

Submitted 12 November, 2021; v1 submitted 29 October, 2020; originally announced October 2020.

Universal S-matrix correlations for complex scattering of many-body wavepackets: theory, simulation and experiment

Authors: Andreas Bereczuk, Barbara Dietz, Jiongning Che, Jack Kuipers, Juan-Diego Urbina, Klaus Richter

Abstract: We present an in-depth study of the universal correlations of scattering-matrix entries required in the framework of non-stationary many-body scattering where the incoming states are localized wavepackets. Contrary to the stationary case the emergence of universal signatures of chaotic dynamics in dynamical observables manifests itself in the emergence of universal correlations of the scattering m… ▽ More We present an in-depth study of the universal correlations of scattering-matrix entries required in the framework of non-stationary many-body scattering where the incoming states are localized wavepackets. Contrary to the stationary case the emergence of universal signatures of chaotic dynamics in dynamical observables manifests itself in the emergence of universal correlations of the scattering matrix at different energies. We use a semiclassical theory based on interfering paths, numerical wave function based simulations and numerical averaging over random-matrix ensembles to calculate such correlations and compare with experimental measurements in microwave graphs, finding excellent agreement. Our calculations show that the universality of the correlators survives the extreme limit of few open channels relevant for electron quantum optics, albeit at the price of dealing with large-cancellation effects requiring the computation of a large class of semiclassical diagrams. △ Less

Submitted 14 October, 2020; originally announced October 2020.

Comments: 9 pages, 10 figures

Journal ref: Phys. Rev. E 103, 052209 (2021)

The Variance of Causal Effect Estimators for Binary V-structures

Authors: Jack Kuipers, Giusi Moffa

Abstract: Adjusting for covariates is a well established method to estimate the total causal effect of an exposure variable on an outcome of interest. Depending on the causal structure of the mechanism under study there may be different adjustment sets, equally valid from a theoretical perspective, leading to identical causal effects. However, in practice, with finite data, estimators built on different set… ▽ More Adjusting for covariates is a well established method to estimate the total causal effect of an exposure variable on an outcome of interest. Depending on the causal structure of the mechanism under study there may be different adjustment sets, equally valid from a theoretical perspective, leading to identical causal effects. However, in practice, with finite data, estimators built on different sets may display different precision. To investigate the extent of this variability we consider the simplest non-trivial non-linear model of a v-structure on three nodes for binary data. We explicitly compute and compare the variance of the two possible different causal estimators. Further, by going beyond leading order asymptotics we show that there are parameter regimes where the set with the asymptotically optimal variance does depend on the edge coefficients, a result which is not captured by the recent leading order developments for general causal models. As a practical consequence, the adjustment set selection needs to account for the relative magnitude of the relationships between variables with respect to the sample size, and cannot rely on purely graphical criteria. △ Less

Submitted 26 April, 2021; v1 submitted 20 April, 2020; originally announced April 2020.

Comments: 15 pages, 2 figures

Efficient Sampling and Structure Learning of Bayesian Networks

Authors: Jack Kuipers, Polina Suter, Giusi Moffa

Abstract: Bayesian networks are probabilistic graphical models widely employed to understand dependencies in high dimensional data, and even to facilitate causal discovery. Learning the underlying network structure, which is encoded as a directed acyclic graph (DAG) is highly challenging mainly due to the vast number of possible networks in combination with the acyclicity constraint. Efforts have focussed o… ▽ More Bayesian networks are probabilistic graphical models widely employed to understand dependencies in high dimensional data, and even to facilitate causal discovery. Learning the underlying network structure, which is encoded as a directed acyclic graph (DAG) is highly challenging mainly due to the vast number of possible networks in combination with the acyclicity constraint. Efforts have focussed on two fronts: constraint-based methods that perform conditional independence tests to exclude edges and score and search approaches which explore the DAG space with greedy or MCMC schemes. Here we synthesise these two fields in a novel hybrid method which reduces the complexity of MCMC approaches to that of a constraint-based method. Individual steps in the MCMC scheme only require simple table lookups so that very long chains can be efficiently obtained. Furthermore, the scheme includes an iterative procedure to correct for errors from the conditional independence tests. The algorithm offers markedly superior performance to alternatives, particularly because DAGs can also be sampled from the posterior distribution, enabling full Bayesian model averaging for much larger Bayesian networks. △ Less

Submitted 9 November, 2021; v1 submitted 21 March, 2018; originally announced March 2018.

Comments: Revised version. 41 pages including 17 pages of supplement, 5 figures and 15 supplemental figures; R package BiDAG is available at https://CRAN.R-project.org/package=BiDAG; Code to reproduce the continuous simulations is available at https://github.com/jackkuipers/esslbn_sim

The Influence of Canalization on the Robustness of Boolean Networks

Authors: Claus Kadelka, Jack Kuipers, Reinhard Laubenbacher

Abstract: Time- and state-discrete dynamical systems are frequently used to model molecular networks. This paper provides a collection of mathematical and computational tools for the study of robustness in Boolean network models. The focus is on networks governed by $k$-canalizing functions, a recently introduced class of Boolean functions that contains the well-studied class of nested canalizing functions.… ▽ More Time- and state-discrete dynamical systems are frequently used to model molecular networks. This paper provides a collection of mathematical and computational tools for the study of robustness in Boolean network models. The focus is on networks governed by $k$-canalizing functions, a recently introduced class of Boolean functions that contains the well-studied class of nested canalizing functions. The activities and sensitivity of a function quantify the impact of input changes on the function output. This paper generalizes the latter concept to $c$-sensitivity and provides formulas for the activities and $c$-sensitivity of general $k$-canalizing functions as well as canalizing functions with more precisely defined structure. A popular measure for the robustness of a network, the Derrida value, can be expressed as a weighted sum of the $c$-sensitivities of the governing canalizing functions, and can also be calculated for a stochastic extension of Boolean networks. These findings provide a computationally efficient way to obtain Derrida values of Boolean networks, deterministic or stochastic, that does not involve simulation. △ Less

Submitted 18 November, 2016; v1 submitted 15 July, 2016; originally announced July 2016.

Comments: 16 pages, 2 figures, 3 tables

Journal ref: Physica D: Nonlinear Phenomena 353 (2017): 39-47

Elastic Instabilities in Flows through Pillared Micro channels

Authors: S. De, J. van der Schaaf, N. G. Deen, J. A. M. Kuipers, E. A. J. F. Peters, J. T. Padding

Abstract: Viscoelastic fluids exhibit elastic instabilities in simple shear flow and flow through curved streamlines. Surprisingly, we found in a porous medium such fluids show strikingly different hydrodynamic instabilities depicted by very large sideways excursions and presence of fast and slow moving lanes which have not been reported before. Particle image velocimetry (PIV) measurements through a pillar… ▽ More Viscoelastic fluids exhibit elastic instabilities in simple shear flow and flow through curved streamlines. Surprisingly, we found in a porous medium such fluids show strikingly different hydrodynamic instabilities depicted by very large sideways excursions and presence of fast and slow moving lanes which have not been reported before. Particle image velocimetry (PIV) measurements through a pillared microchannel, provide experimental evidence of such instabilities at very low Reynolds number (< 0.01). We observe a transition from a symmetric laminar to an asymmetric flow, which finally transforms to a nonlinear aperiodic flow with strong lateral movements. The instability is characterized by a rapid increase in spatial and temporal fluctuations of velocity components and pressure at a critical Deborah number (De). Our experiments reveal the presence of a fascinating interplay between pore space and fluid rheology. △ Less

Submitted 13 July, 2016; originally announced July 2016.

Comments: 11 pages, 8 figures

From Many-Particle Interference to Correlation Spectroscopy

Authors: Mattia Walschaers, Jack Kuipers, Andreas Buchleitner

Abstract: We show how robust statistical features of a many-particle quantum state's two-point correlations after transmission through a multi-mode random scatterer can be used as a sensitive probe of the injected particles' mutual indistinguishability. This generalises Hong-Ou-Mandel interference as a diagnostic tool for many-particle transmission signals across multi-mode random scatterers. Furthermore, w… ▽ More We show how robust statistical features of a many-particle quantum state's two-point correlations after transmission through a multi-mode random scatterer can be used as a sensitive probe of the injected particles' mutual indistinguishability. This generalises Hong-Ou-Mandel interference as a diagnostic tool for many-particle transmission signals across multi-mode random scatterers. Furthermore, we show how, from such statistical features of the many-particle interference pattern, information can be deduced on the temporal structure of the many-particle input state, by inspection of the many-particle interference with an additional probe particle of tuneable distinguishability. △ Less

Submitted 5 September, 2016; v1 submitted 3 April, 2016; originally announced April 2016.

Journal ref: Phys. Rev. A 94, 020104 (2016)

Partition MCMC for inference on acyclic digraphs

Authors: Jack Kuipers, Giusi Moffa

Abstract: Acyclic digraphs are the underlying representation of Bayesian networks, a widely used class of probabilistic graphical models. Learning the underlying graph from data is a way of gaining insights about the structural properties of a domain. Structure learning forms one of the inference challenges of statistical graphical models. MCMC methods, notably structure MCMC, to sample graphs from the po… ▽ More Acyclic digraphs are the underlying representation of Bayesian networks, a widely used class of probabilistic graphical models. Learning the underlying graph from data is a way of gaining insights about the structural properties of a domain. Structure learning forms one of the inference challenges of statistical graphical models. MCMC methods, notably structure MCMC, to sample graphs from the posterior distribution given the data are probably the only viable option for Bayesian model averaging. Score modularity and restrictions on the number of parents of each node allow the graphs to be grouped into larger collections, which can be scored as a whole to improve the chain's convergence. Current examples of algorithms taking advantage of grou** are the biased order MCMC, which acts on the alternative space of permuted triangular matrices, and non ergodic edge reversal moves. Here we propose a novel algorithm, which employs the underlying combinatorial structure of DAGs to define a new grou**. As a result convergence is improved compared to structure MCMC, while still retaining the property of producing an unbiased sample. Finally the method can be combined with edge reversal moves to improve the sampler further. △ Less

Submitted 20 October, 2015; v1 submitted 20 April, 2015; originally announced April 2015.

Comments: Revised version. 34 pages, 16 figures. R code available at https://github.com/annlia/partitionMCMC

Journal ref: J. Am. Stat. Assoc. 112 (2017) 282-299

Multistate Nested Canalizing Functions and Their Networks

Authors: Claus Kadelka, Yuan Li, Jack Kuipers, John O. Adeyeye, Reinhard Laubenbacher

Abstract: This paper provides a collection of mathematical and computational tools for the study of robustness in nonlinear gene regulatory networks, represented by time- and state-discrete dynamical systems taking on multiple states. The focus is on networks governed by nested canalizing functions (NCFs), first introduced in the Boolean context by S. Kauffman. After giving a general definition of NCFs we a… ▽ More This paper provides a collection of mathematical and computational tools for the study of robustness in nonlinear gene regulatory networks, represented by time- and state-discrete dynamical systems taking on multiple states. The focus is on networks governed by nested canalizing functions (NCFs), first introduced in the Boolean context by S. Kauffman. After giving a general definition of NCFs we analyze the class of such functions. We derive a formula for the normalized average $c$-sensitivities of multistate NCFs, which enables the calculation of the Derrida plot, a popular measure of network stability. We also provide a unique canonical parametrized polynomial form of NCFs. This form has several consequences. We can easily generate NCFs for varying parameter choices, and derive a closed form formula for the number of such functions in a given number of variables, as well as an asymptotic formula. Finally, we compute the number of equivalence classes of NCFs under permutation of variables. Together, the results of the paper represent a useful mathematical framework for the study of NCFs and their dynamic networks. △ Less

Submitted 29 August, 2016; v1 submitted 14 November, 2014; originally announced November 2014.

Comments: 15 pages, 1 figure. arXiv admin note: substantial text overlap with arXiv:1304.3707

A Statistical Benchmark for BosonSampling

Authors: Mattia Walschaers, Jack Kuipers, Juan-Diego Urbina, Klaus Mayer, Malte C. Tichy, Klaus Richter, Andreas Buchleitner

Abstract: Computing the state of a quantum mechanical many-body system composed of indistinguishable particles distributed over a multitude of modes is one of the paradigmatic test cases of computational complexity theory: Beyond well-understood quantum statistical effects, the coherent superposition of many-particle amplitudes rapidly overburdens classical computing devices - essentially by creating extrem… ▽ More Computing the state of a quantum mechanical many-body system composed of indistinguishable particles distributed over a multitude of modes is one of the paradigmatic test cases of computational complexity theory: Beyond well-understood quantum statistical effects, the coherent superposition of many-particle amplitudes rapidly overburdens classical computing devices - essentially by creating extremely complicated interference patterns, which also challenge experimental resolution. With the advent of controlled many-particle interference experiments, optical set-ups that can efficiently probe many-boson wave functions - baptised BosonSamplers - have therefore been proposed as efficient quantum simulators which outperform any classical computing device, and thereby challenge the extended Church-Turing thesis, one of the fundamental dogmas of computer science. However, as in all experimental quantum simulations of truly complex systems, there remains one crucial problem: How to certify that a given experimental measurement record is an unambiguous result of sampling bosons rather than fermions or distinguishable particles, or of uncontrolled noise? In this contribution, we describe a statistical signature of many-body quantum interference, which can be used as an experimental (and classically computable) benchmark for BosonSampling. △ Less

Submitted 30 October, 2014; originally announced October 2014.

Comments: 9 pages, 5 figures

Journal ref: New J. Phys. 18, 032001 (2016)

Multiparticle correlations in mesoscopic scattering: boson sampling, birthday paradox, and Hong-Ou-Mandel profiles

Authors: Juan-Diego Urbina, Jack Kuipers, Quirin Hummel, Sho Matsumoto, Klaus Richter

Abstract: The interplay between single-particle interference and quantum indistinguishability leads to signature correlations in many-body scattering. We uncover these with a semiclassical calculation of the transmission probabilities through mesoscopic cavities for systems of non-interacting particles. For chaotic cavities we provide the universal form of the first two moments of the transmission probabili… ▽ More The interplay between single-particle interference and quantum indistinguishability leads to signature correlations in many-body scattering. We uncover these with a semiclassical calculation of the transmission probabilities through mesoscopic cavities for systems of non-interacting particles. For chaotic cavities we provide the universal form of the first two moments of the transmission probabilities over ensembles of random unitary matrices, including weak localization and dephasing effects. If the incoming many-body state consists of two macroscopically occupied wavepackets, their time delay drives a quantum-classical transition along a boundary determined by the bosonic birthday paradox. Mesoscopic chaotic scattering of Bose-Einstein condensates is then a realistic candidate to build a boson sampler and to observe the macroscopic Hong-Ou-Mandel effect. △ Less

Submitted 14 April, 2016; v1 submitted 4 September, 2014; originally announced September 2014.

Comments: 6+11 pages, 3+3 figures

Journal ref: Phys. Rev. Lett. 116, 100401 (2016)

Efficient semiclassical approach for time delays

Authors: Jack Kuipers, Dmitry V. Savin, Martin Sieber

Abstract: The Wigner time delay, defined by the energy derivative of the total scattering phase shift, is an important spectral measure of an open quantum system characterising the duration of the scattering event. It is related to the trace of the Wigner-Smith matrix Q that also encodes other time-delay characteristics. For chaotic cavities, these exhibit universal fluctuations that are commonly described… ▽ More The Wigner time delay, defined by the energy derivative of the total scattering phase shift, is an important spectral measure of an open quantum system characterising the duration of the scattering event. It is related to the trace of the Wigner-Smith matrix Q that also encodes other time-delay characteristics. For chaotic cavities, these exhibit universal fluctuations that are commonly described within random matrix theory. Here, we develop a new semiclassical approach to the time-delay matrix which is formulated in terms of the classical trajectories that connect the exterior and interior regions of the system. This approach is superior to previous treatments because it avoids the energy derivative. We demonstrate the method's efficiency by going beyond previous work in studying the time-delay statistics for chaotic cavities with perfectly connected leads. In particular, the universality for moment generating functions of the proper time-delays (eigenvalues of Q) is established up to third order in the inverse number of scattering channels for systems with and without time-reversal symmetry. Semiclassical results are then obtained for a further two orders. We also show the equivalence of random matrix and semiclassical results for the second moments and for the variance of the Wigner time delay at any channel number. △ Less

Submitted 11 December, 2014; v1 submitted 4 September, 2014; originally announced September 2014.

Comments: 50 pages, 8 figures

Journal ref: New J. Phys. 16, 123018 (2014)

Existence of Secure Equilibrium in Multi-Player Games with Perfect Information

Authors: Julie De Pril, János Flesch, Jeroen Kuipers, Gijs Schoenmakers, Koos Vrieze

Abstract: Secure equilibrium is a refinement of Nash equilibrium, which provides some security to the players against deviations when a player changes his strategy to another best response strategy. The concept of secure equilibrium is specifically developed for assume-guarantee synthesis and has already been applied in this context. Yet, not much is known about its existence in games with more than two pla… ▽ More Secure equilibrium is a refinement of Nash equilibrium, which provides some security to the players against deviations when a player changes his strategy to another best response strategy. The concept of secure equilibrium is specifically developed for assume-guarantee synthesis and has already been applied in this context. Yet, not much is known about its existence in games with more than two players. In this paper, we establish the existence of secure equilibrium in two classes of multi-player perfect information turn-based games: (1) in games with possibly probabilistic transitions, having countable state and finite action spaces and bounded and continuous payoff functions, and (2) in games with only deterministic transitions, having arbitrary state and action spaces and Borel payoff functions with a finite range (in particular, qualitative Borel payoff functions). We show that these results apply to several types of games studied in the literature. △ Less

Submitted 7 May, 2014; originally announced May 2014.

Comments: 16 pages

Addendum on the scoring of Gaussian directed acyclic graphical models

Authors: Jack Kuipers, Giusi Moffa, David Heckerman

Abstract: We provide a correction to the expression for scoring Gaussian directed acyclic graphical models derived in Geiger and Heckerman [Ann. Statist. 30 (2002) 1414-1440] and discuss how to evaluate the score efficiently. We provide a correction to the expression for scoring Gaussian directed acyclic graphical models derived in Geiger and Heckerman [Ann. Statist. 30 (2002) 1414-1440] and discuss how to evaluate the score efficiently. △ Less

Submitted 17 May, 2021; v1 submitted 27 February, 2014; originally announced February 2014.

Comments: Published in at http://dx.doi.org/10.1214/14-AOS1217 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org). Typo in definition of R and one sentence corrected

Report number: IMS-AOS-AOS1217

Journal ref: Annals of Statistics 2014, Vol. 42, No. 4, 1689-1691

Code Optimization in FORM

Authors: J. Kuipers, T. Ueda, J. A. M. Vermaseren

Abstract: We describe the implementation of output code optimization in the open source computer algebra system FORM. This implementation is based on recently discovered techniques of Monte Carlo tree search to find efficient multivariate Horner schemes, in combination with other optimization algorithms, such as common subexpression elimination. For systems for which no specific knowledge is provided it per… ▽ More We describe the implementation of output code optimization in the open source computer algebra system FORM. This implementation is based on recently discovered techniques of Monte Carlo tree search to find efficient multivariate Horner schemes, in combination with other optimization algorithms, such as common subexpression elimination. For systems for which no specific knowledge is provided it performs significantly better than other methods we could compare with. Because the method has a number of free parameters, we also show some methods by which to tune them to different types of problems. △ Less

Submitted 25 October, 2013; originally announced October 2013.

Comments: 31 pages

Report number: Nikhef 2013-036, TTP13-031, SFB/CPP-13-80 MSC Class: 97Rxx

Quantum graphs whose spectra mimic the zeros of the Riemann zeta function

Authors: Jack Kuipers, Quirin Hummel, Klaus Richter

Abstract: One of the most famous problems in mathematics is the Riemann hypothesis: that the non-trivial zeros of the Riemann zeta function lie on a line in the complex plane. One way to prove the hypothesis would be to identify the zeros as eigenvalues of a Hermitian operator, many of whose properties can be derived through the analogy to quantum chaos. Using this, we construct a set of quantum graphs that… ▽ More One of the most famous problems in mathematics is the Riemann hypothesis: that the non-trivial zeros of the Riemann zeta function lie on a line in the complex plane. One way to prove the hypothesis would be to identify the zeros as eigenvalues of a Hermitian operator, many of whose properties can be derived through the analogy to quantum chaos. Using this, we construct a set of quantum graphs that have the same oscillating part of the density of states as the Riemann zeros, offering an explanation of the overall minus sign. The smooth part is completely different, and hence also the spectrum, but the graphs pick out the low-lying zeros. △ Less

Submitted 26 February, 2014; v1 submitted 23 July, 2013; originally announced July 2013.

Comments: 8 pages, 8 pdf figures

Journal ref: Phys. Rev. Lett 112, 070406 (2014)

Combinatorial theory of the semiclassical evaluation of transport moments II: Algorithmic approach for moment generating functions

Authors: G. Berkolaiko, J. Kuipers

Abstract: Electronic transport through chaotic quantum dots exhibits universal behaviour which can be understood through the semiclassical approximation. Within the approximation, transport moments reduce to codifying classical correlations between scattering trajectories. These can be represented as ribbon graphs and we develop an algorithmic combinatorial method to generate all such graphs with a given ge… ▽ More Electronic transport through chaotic quantum dots exhibits universal behaviour which can be understood through the semiclassical approximation. Within the approximation, transport moments reduce to codifying classical correlations between scattering trajectories. These can be represented as ribbon graphs and we develop an algorithmic combinatorial method to generate all such graphs with a given genus. This provides an expansion of the linear transport moments for systems both with and without time reversal symmetry. The computational implementation is then able to progress several orders higher than previous semiclassical formulae as well as those derived from an asymptotic expansion of random matrix results. The patterns observed also suggest a general form for the higher orders. △ Less

Submitted 19 November, 2013; v1 submitted 11 July, 2013; originally announced July 2013.

Comments: 37 pages; 19 figures

MSC Class: 81V65; 81Q20; 15B52; 81Q50

Journal ref: J. Math. Phys. 54, 123505 (2013)

Combinatorial theory of the semiclassical evaluation of transport moments I: Equivalence with the random matrix approach

Authors: G. Berkolaiko, J. Kuipers

Abstract: To study electronic transport through chaotic quantum dots, there are two main theoretical approachs. One involves substituting the quantum system with a random scattering matrix and performing appropriate ensemble averaging. The other treats the transport in the semiclassical approximation and studies correlations among sets of classical trajectories. There are established evaluation procedures w… ▽ More To study electronic transport through chaotic quantum dots, there are two main theoretical approachs. One involves substituting the quantum system with a random scattering matrix and performing appropriate ensemble averaging. The other treats the transport in the semiclassical approximation and studies correlations among sets of classical trajectories. There are established evaluation procedures within the semiclassical evaluation that, for several linear and non-linear transport moments to which they were applied, have always resulted in the agreement with random matrix predictions. We prove that this agreement is universal: any semiclassical evaluation within the accepted procedures is equivalent to the evaluation within random matrix theory. The equivalence is shown by develo** a combinatorial interpretation of the trajectory sets as ribbon graphs (maps) with certain properties and exhibiting systematic cancellations among their contributions. Remaining trajectory sets can be identified with primitive (palindromic) factorisations whose number gives the coefficients in the corresponding expansion of the moments of random matrices. The equivalence is proved for systems with and without time reversal symmetry. △ Less

Submitted 20 September, 2013; v1 submitted 21 May, 2013; originally announced May 2013.

Comments: 30 pages, 13 figures; added some examples

MSC Class: 81V65; 81Q20; 15B52; 81Q50

Journal ref: J. Math. Phys. 54, 112103 (2013)

Transport moments and Andreev billiards with tunnel barriers

Authors: Jack Kuipers, Klaus Richter

Abstract: Open chaotic systems are expected to possess universal transport statistics and recently there have been many advances in understanding and obtaining expressions for their transport moments. However when tunnel barriers are added, which represents the situation in more general experimental physical systems much less is known about the behaviour of the moments. By incorporating tunnel barriers in t… ▽ More Open chaotic systems are expected to possess universal transport statistics and recently there have been many advances in understanding and obtaining expressions for their transport moments. However when tunnel barriers are added, which represents the situation in more general experimental physical systems much less is known about the behaviour of the moments. By incorporating tunnel barriers in the recursive semiclassical diagrammatic approach we obtain the moment generating function of the transmission eigenvalues at leading and subleading order. For reflection quantities quantum mechanical tunneling phases play an essential role and we introduce new structures to deal with them. This allows us to obtain the moment generating function of the reflection eigenvalues and the Wigner delay times at leading order. Our semiclassical results are in complementary regimes to the leading order results derived from random matrix theory expanding the range of theoretically known moments. As a further application we derive to leading order the density of states of Andreev billiards coupled to a superconductor through tunnel barriers. △ Less

Submitted 12 February, 2013; v1 submitted 30 October, 2012; originally announced October 2012.

Comments: Final refereed version. 25 pages, 7 figures. Corrected a reference

Journal ref: J Phys A 46 (2013) 055101

Improving multivariate Horner schemes with Monte Carlo tree search

Authors: J. Kuipers, J. A. M. Vermaseren, A. Plaat, H. J. van den Herik

Abstract: Optimizing the cost of evaluating a polynomial is a classic problem in computer science. For polynomials in one variable, Horner's method provides a scheme for producing a computationally efficient form. For multivariate polynomials it is possible to generalize Horner's method, but this leaves freedom in the order of the variables. Traditionally, greedy schemes like most-occurring variable first a… ▽ More Optimizing the cost of evaluating a polynomial is a classic problem in computer science. For polynomials in one variable, Horner's method provides a scheme for producing a computationally efficient form. For multivariate polynomials it is possible to generalize Horner's method, but this leaves freedom in the order of the variables. Traditionally, greedy schemes like most-occurring variable first are used. This simple textbook algorithm has given remarkably efficient results. Finding better algorithms has proved difficult. In trying to improve upon the greedy scheme we have implemented Monte Carlo tree search, a recent search method from the field of artificial intelligence. This results in better Horner schemes and reduces the cost of evaluating polynomials, sometimes by factors up to two. △ Less

Submitted 30 July, 2012; originally announced July 2012.

Comments: 5 pages

FORM version 4.0

Authors: J. Kuipers, T. Ueda, J. A. M. Vermaseren, J. Vollinga

Abstract: We present version 4.0 of the symbolic manipulation system FORM. The most important new features are manipulation of rational polynomials and the factorization of expressions. Many other new functions and commands are also added; some of them are very general, while others are designed for building specific high level packages, such as one for Groebner bases. New is also the checkpoint facility, t… ▽ More We present version 4.0 of the symbolic manipulation system FORM. The most important new features are manipulation of rational polynomials and the factorization of expressions. Many other new functions and commands are also added; some of them are very general, while others are designed for building specific high level packages, such as one for Groebner bases. New is also the checkpoint facility, that allows for periodic backups during long calculations. Lastly, FORM 4.0 has become available as open source under the GNU General Public License version 3. △ Less

Submitted 29 March, 2012; originally announced March 2012.

Comments: 26 pages. Uses axodraw

Uniform random generation of large acyclic digraphs

Authors: Jack Kuipers, Giusi Moffa

Abstract: Directed acyclic graphs are the basic representation of the structure underlying Bayesian networks, which represent multivariate probability distributions. In many practical applications, such as the reverse engineering of gene regulatory networks, not only the estimation of model parameters but the reconstruction of the structure itself is of great interest. As well as for the assessment of diffe… ▽ More Directed acyclic graphs are the basic representation of the structure underlying Bayesian networks, which represent multivariate probability distributions. In many practical applications, such as the reverse engineering of gene regulatory networks, not only the estimation of model parameters but the reconstruction of the structure itself is of great interest. As well as for the assessment of different structure learning algorithms in simulation studies, a uniform sample from the space of directed acyclic graphs is required to evaluate the prevalence of certain structural features. Here we analyse how to sample acyclic digraphs uniformly at random through recursive enumeration, an approach previously thought too computationally involved. Based on complexity considerations, we discuss in particular how the enumeration directly provides an exact method, which avoids the convergence issues of the alternative Markov chain methods and is actually computationally much faster. The limiting behaviour of the distribution of acyclic digraphs then allows us to sample arbitrarily large graphs. Building on the ideas of recursive enumeration based sampling we also introduce a novel hybrid Markov chain with much faster convergence than current alternatives while still being easy to adapt to various restrictions. Finally we discuss how to include such restrictions in the combinatorial enumeration and the new hybrid Markov chain method for efficient uniform sampling of the corresponding graphs. △ Less

Submitted 14 November, 2013; v1 submitted 29 February, 2012; originally announced February 2012.

Comments: 15 pages, 2 figures. To appear in Statistics and Computing

Universality in chaotic quantum transport: The concordance between random matrix and semiclassical theories

Authors: Gregory Berkolaiko, Jack Kuipers

Abstract: Electronic transport through chaotic quantum dots exhibits universal, system independent, properties, consistent with random matrix theory. The quantum transport can also be rooted, via the semiclassical approximation, in sums over the classical scattering trajectories. Correlations between such trajectories can be organized diagrammatically and have been shown to yield universal answers for some… ▽ More Electronic transport through chaotic quantum dots exhibits universal, system independent, properties, consistent with random matrix theory. The quantum transport can also be rooted, via the semiclassical approximation, in sums over the classical scattering trajectories. Correlations between such trajectories can be organized diagrammatically and have been shown to yield universal answers for some observables. Here, we develop the general combinatorial treatment of the semiclassical diagrams, through a connection to factorizations of permutations. We show agreement between the semiclassical and random matrix approaches to the moments of the transmission eigenvalues. The result is valid for all moments to all orders of the expansion in inverse channel number for all three main symmetry classes (with and without time reversal symmetry and spin-orbit interaction) and extends to nonlinear statistics. This finally explains the applicability of random matrix theory to chaotic quantum transport in terms of the underlying dynamics as well as providing semiclassical access to the probability density of the transmission eigenvalues. △ Less

Submitted 4 May, 2012; v1 submitted 21 November, 2011; originally announced November 2011.

Comments: Refereed version. 5 pages, 4 figures

Journal ref: Phys Rev E 85 (2012) 045201(R)

Conductance fluctuations in chaotic systems with tunnel barriers

Authors: Daniel Waltner, Jack Kuipers, Philippe Jacquod, Klaus Richter

Abstract: Quantum effects are expected to disappear in the short-wavelength, semiclassical limit. As a matter of fact, recent investigations of transport through quantum chaotic systems have demonstrated the exponential suppression of the weak localization corrections to the conductance and of the Fano factor for shot-noise when the Ehrenfest time exceeds the electronic dwell time. On the other hand, conduc… ▽ More Quantum effects are expected to disappear in the short-wavelength, semiclassical limit. As a matter of fact, recent investigations of transport through quantum chaotic systems have demonstrated the exponential suppression of the weak localization corrections to the conductance and of the Fano factor for shot-noise when the Ehrenfest time exceeds the electronic dwell time. On the other hand, conductance fluctuations, an effect of quantum coherence, retain their universal value in the limit of the ratio of Ehrenfest time over dwell time to infinity, when the system is ideally coupled to external leads. Motivated by this intriguing result we investigate conductance fluctuations through quantum chaotic cavities coupled to external leads via (tunnel) barriers of arbitrary transparency. Using the trajectory-based semiclassical theory of transport, we find a linear Ehrenfest time-dependence of the conductance variance showing a nonmonotonous, sinusoidal behavior as a function of the transperancy. Most notably, we find an increase of the conductance fluctuations with the Ehrenfest time, above their universal value, for the transparency less than 0.5. These results, confirmed by numerical simulations, show that, contrarily to the common wisdom, effects of quantum coherence may increase in the semiclassical limit, under special circumstances. △ Less

Submitted 25 August, 2011; originally announced August 2011.

Journal ref: Phys. Rev. B 85 (2012) 024302

Sequential Monte Carlo EM for multivariate probit models

Authors: Giusi Moffa, Jack Kuipers

Abstract: Multivariate probit models (MPM) have the appealing feature of capturing some of the dependence structure between the components of multidimensional binary responses. The key for the dependence modelling is the covariance matrix of an underlying latent multivariate Gaussian. Most approaches to MLE in multivariate probit regression rely on MCEM algorithms to avoid computationally intensive evaluati… ▽ More Multivariate probit models (MPM) have the appealing feature of capturing some of the dependence structure between the components of multidimensional binary responses. The key for the dependence modelling is the covariance matrix of an underlying latent multivariate Gaussian. Most approaches to MLE in multivariate probit regression rely on MCEM algorithms to avoid computationally intensive evaluations of multivariate normal orthant probabilities. As an alternative to the much used Gibbs sampler a new SMC sampler for truncated multivariate normals is proposed. The algorithm proceeds in two stages where samples are first drawn from truncated multivariate Student $t$ distributions and then further evolved towards a Gaussian. The sampler is then embedded in a MCEM algorithm. The sequential nature of SMC methods can be exploited to design a fully sequential version of the EM, where the samples are simply updated from one iteration to the next rather than resampled from scratch. Recycling the samples in this manner significantly reduces the computational cost. An alternative view of the standard conditional maximisation step provides the basis for an iterative procedure to fully perform the maximisation needed in the EM algorithm. The identifiability of MPM is also thoroughly discussed. In particular, the likelihood invariance can be embedded in the EM algorithm to ensure that constrained and unconstrained maximisation are equivalent. A simple iterative procedure is then derived for either maximisation which takes effectively no computational time. The method is validated by applying it to the widely analysed Six Cities dataset and on a higher dimensional simulated example. Previous approaches to the Six Cities overly restrict the parameter space but, by considering the correct invariance, the maximum likelihood is quite naturally improved when treating the full unrestricted model. △ Less

Submitted 14 November, 2013; v1 submitted 12 July, 2011; originally announced July 2011.

Comments: 26 pages, 2 figures. In press, Computational Statistics & Data Analysis

About a conjectured basis for Multiple Zeta Values

Authors: J. Kuipers, J. A. M. Vermaseren

Abstract: We confirm a conjecture about the construction of basis elements for the multiple zeta values (MZVs) at weight 27 and weight 28. Both show as expected one element that is twofold extended. This is done with some lengthy computer algebra calculations using TFORM to determine explicit bases for the MZVs at these weights. We confirm a conjecture about the construction of basis elements for the multiple zeta values (MZVs) at weight 27 and weight 28. Both show as expected one element that is twofold extended. This is done with some lengthy computer algebra calculations using TFORM to determine explicit bases for the MZVs at these weights. △ Less

Submitted 10 May, 2011; originally announced May 2011.

Comments: 6 pages

Report number: Nikhef 11-008

Transport moments beyond the leading order

Authors: Gregory Berkolaiko, Jack Kuipers

Abstract: For chaotic cavities with scattering leads attached, transport properties can be approximated in terms of the classical trajectories which enter and exit the system. With a semiclassical treatment involving fine correlations between such trajectories we develop a diagrammatic technique to calculate the moments of various transport quantities. Namely, we find the moments of the transmission and ref… ▽ More For chaotic cavities with scattering leads attached, transport properties can be approximated in terms of the classical trajectories which enter and exit the system. With a semiclassical treatment involving fine correlations between such trajectories we develop a diagrammatic technique to calculate the moments of various transport quantities. Namely, we find the moments of the transmission and reflection eigenvalues for systems with and without time reversal symmetry. We also derive related quantities involving an energy dependence: the moments of the Wigner delay times and the density of states of chaotic Andreev billiards, where we find that the gap in the density persists when subleading corrections are included. Finally, we show how to adapt our techniques to non-linear statistics by calculating the correlation between transport moments. In each setting, the answer for the $n$-th moment is obtained for arbitrary $n$ (in the form of a moment generating function) and for up to the three leading orders in terms of the inverse channel number. Our results suggest patterns which should hold for further corrections and by matching with the low order moments available from random matrix theory we derive likely higher order generating functions. △ Less

Submitted 11 July, 2011; v1 submitted 16 December, 2010; originally announced December 2010.

Comments: Refereed version. 43 pages, 10 figures

Journal ref: New J Phys 13 (2011) 063020

Conductance and Thermopower of Ballistic Andreev Cavities

Authors: Thomas Engl, Jack Kuipers, Klaus Richter

Abstract: When coupling a superconductor to a normal conducting region the physical properties of the system are highly affected by the superconductor. We will investigate the effect of one or two superconductors on the conductance of a ballistic chaotic quantum dot to leading order in the total channel number using trajectory based semiclassics. The results show that the effect of one superconductor on the… ▽ More When coupling a superconductor to a normal conducting region the physical properties of the system are highly affected by the superconductor. We will investigate the effect of one or two superconductors on the conductance of a ballistic chaotic quantum dot to leading order in the total channel number using trajectory based semiclassics. The results show that the effect of one superconductor on the conductance is of the order of the number of channels and that the sign of the correction from the Drude conductance depends on the particular ratios of the numbers of channels of the superconducting and normal conducting leads. In the case of two superconductors with the same chemical potential we will also see how the conductance and the sign of quantum corrections are affected by their phase difference. As far as random matrix theory results exist these are reproduced by our calculations. Furthermore in the case that the chemical potential of the superconductors is the same as that of one of the two normal leads the conductance shows, under certain conditions, similar effects as a normal metal-superconductor junction. The semiclassical framework is also able to treat the thermopower of chaotic Andreev billiards consisting of one chaotic dot, two normal leads and two superconducting islands and shows it to be antisymmetric in the phase difference of the superconductors. △ Less

Submitted 10 January, 2011; v1 submitted 15 December, 2010; originally announced December 2010.

Comments: 23 pages

Journal ref: Phys. Rev. B 83, 205414 (2011)

Weak localization in mesoscopic hole transport: Berry phases and classical correlations

Authors: Viktor Krueckl, Michael Wimmer, Inanc Adagideli, Jack Kuipers, Klaus Richter

Abstract: We consider phase-coherent transport through ballistic and diffusive two-dimensional hole systems based on the Kohn-Luttinger Hamiltonian. We show that intrinsic heavy-hole light-hole coupling gives rise to clear-cut signatures of an associated Berry phase in the weak localization which renders the magneto-conductance profile distinctly different from electron transport. Non-universal classical co… ▽ More We consider phase-coherent transport through ballistic and diffusive two-dimensional hole systems based on the Kohn-Luttinger Hamiltonian. We show that intrinsic heavy-hole light-hole coupling gives rise to clear-cut signatures of an associated Berry phase in the weak localization which renders the magneto-conductance profile distinctly different from electron transport. Non-universal classical correlations determine the strength of these Berry phase effects and the effective symmetry class, leading even to antilocalization-type features for circular quantum dots and Aharonov-Bohm rings in the absence of additional spin-orbit interaction. Our semiclassical predictions are quantitatively confirmed by numerical transport calculations. △ Less

Submitted 28 September, 2010; originally announced September 2010.

Journal ref: Phys. Rev. Lett. 106, 146801 (2011)

Ehrenfest-time dependence of counting statistics for chaotic ballistic systems

Authors: Daniel Waltner, Jack Kuipers, Klaus Richter

Abstract: Transport properties of open chaotic ballistic systems and their statistics can be expressed in terms of the scattering matrix connecting incoming and outgoing wavefunctions. Here we calculate the dependence of correlation functions of arbitrarily many pairs of scattering matrices at different energies on the Ehrenfest time using trajectory based semiclassical methods. This enables us to verify th… ▽ More Transport properties of open chaotic ballistic systems and their statistics can be expressed in terms of the scattering matrix connecting incoming and outgoing wavefunctions. Here we calculate the dependence of correlation functions of arbitrarily many pairs of scattering matrices at different energies on the Ehrenfest time using trajectory based semiclassical methods. This enables us to verify the prediction from effective random matrix theory that one part of the correlation function obtains an exponential dam** depending on the Ehrenfest time, while also allowing us to obtain the additional contribution which arises from bands of always correlated trajectories. The resulting Ehrenfest-time dependence, responsible e.g. for secondary gaps in the density of states of Andreev billiards, can also be seen to have strong effects on other transport quantities like the distribution of delay times. △ Less

Submitted 20 May, 2011; v1 submitted 9 July, 2010; originally announced July 2010.

Comments: Refereed version. 15 pages, 14 figures

Journal ref: Phys Rev B 83 (2011) 195315

Ehrenfest time dependence of quantum transport corrections and spectral statistics

Authors: Daniel Waltner, Jack Kuipers

Abstract: The Ehrenfest time scale in quantum transport separates essentially classical propagation from wave interference and here we consider its effect on the transmission and reflection through quantum dots. In particular we calculate the Ehrenfest time dependence of the next-to-leading-order quantum corrections to the transmission and reflection for dc and ac transport and check that our results are co… ▽ More The Ehrenfest time scale in quantum transport separates essentially classical propagation from wave interference and here we consider its effect on the transmission and reflection through quantum dots. In particular we calculate the Ehrenfest time dependence of the next-to-leading-order quantum corrections to the transmission and reflection for dc and ac transport and check that our results are consistent with current conservation relations. Looking as well at closed systems, we finally demonstrate how the contributions analyzed here imply changes in the calculation given in [P. W. Brouwer, S. Rahav and C. Tian, Phys. Rev. E, 74, 066208 (2006)] of the next to leading order of the spectral form factor. Our semiclassical result coincides with the result obtained in [C. Tian and A. I. Larkin, Phys. Rev. B, 70, 035305 (2004)] by field-theoretical methods. △ Less

Submitted 5 April, 2011; v1 submitted 1 June, 2010; originally announced June 2010.

Comments: 11 pages, 9 figures

Journal ref: Physical Review E 82, 066205 (2010)

The density of states of chaotic Andreev billiards

Authors: Jack Kuipers, Thomas Engl, Gregory Berkolaiko, Cyril Petitjean, Daniel Waltner, Klaus Richter

Abstract: Quantum cavities or dots have markedly different properties depending on whether their classical counterparts are chaotic or not. Connecting a superconductor to such a cavity leads to notable proximity effects, particularly the appearance, predicted by random matrix theory, of a hard gap in the excitation spectrum of quantum chaotic systems. Andreev billiards are interesting examples of such struc… ▽ More Quantum cavities or dots have markedly different properties depending on whether their classical counterparts are chaotic or not. Connecting a superconductor to such a cavity leads to notable proximity effects, particularly the appearance, predicted by random matrix theory, of a hard gap in the excitation spectrum of quantum chaotic systems. Andreev billiards are interesting examples of such structures built with superconductors connected to a ballistic normal metal billiard since each time an electron hits the superconducting part it is retroreflected as a hole (and vice-versa). Using a semiclassical framework for systems with chaotic dynamics, we show how this reflection, along with the interference due to subtle correlations between the classical paths of electrons and holes inside the system, are ultimately responsible for the gap formation. The treatment can be extended to include the effects of a symmetry breaking magnetic field in the normal part of the billiard or an Andreev billiard connected to two phase shifted superconductors. Therefore we are able to see how these effects can remold and eventually suppress the gap. Furthermore the semiclassical framework is able to cover the effect of a finite Ehrenfest time which also causes the gap to shrink. However for intermediate values this leads to the appearance of a second hard gap - a clear signature of the Ehrenfest time. △ Less

Submitted 20 May, 2011; v1 submitted 8 April, 2010; originally announced April 2010.

Comments: Refereed version. 23 pages, 19 figures

Journal ref: Phys Rev B 83 (2011) 195316

Quantum Corrections to Fidelity Decay in Chaotic Systems

Authors: Boris Gutkin, Daniel Waltner, Martha Gutierrez, Jack Kuipers, Klaus Richter

Abstract: By considering correlations between classical orbits we derive semiclassical expressions for the decay of the quantum fidelity amplitude for classically chaotic quantum systems, as well as for its squared modulus, the fidelity or Loschmidt echo. Our semiclassical results for the fidelity amplitude agree with random matrix theory (RMT) and supersymmetry predictions in the universal Fermi golden r… ▽ More By considering correlations between classical orbits we derive semiclassical expressions for the decay of the quantum fidelity amplitude for classically chaotic quantum systems, as well as for its squared modulus, the fidelity or Loschmidt echo. Our semiclassical results for the fidelity amplitude agree with random matrix theory (RMT) and supersymmetry predictions in the universal Fermi golden rule regime. The calculated quantum corrections can be viewed as arising from a static random perturbation acting on nearly self-retracing interfering paths, and hence will be suppressed for time-varying perturbations. Moreover, using trajectory-based methods we show a relation, recently obtained in RMT, between the fidelity amplitude and the cross-form factor for parametric level correlations. Beyond RMT, we compute Ehrenfest-time effects on the fidelity amplitude. Furthermore our semiclassical approach allows for a unified treatment of the fidelity, both in the Fermi golden rule and Lyapunov regimes, demonstrating that quantum corrections are suppressed in the latter. △ Less

Submitted 4 November, 2009; originally announced November 2009.

Comments: 14 pages, 4 figures

Moments of the Wigner delay times

Authors: Gregory Berkolaiko, Jack Kuipers

Abstract: The Wigner time delay is a measure of the time spent by a particle inside the scattering region of an open system. For chaotic systems, the statistics of the individual delay times (whose average is the Wigner time delay) are thought to be well described by random matrix theory. Here we present a semiclassical derivation showing the validity of random matrix results. In order to simplify the sem… ▽ More The Wigner time delay is a measure of the time spent by a particle inside the scattering region of an open system. For chaotic systems, the statistics of the individual delay times (whose average is the Wigner time delay) are thought to be well described by random matrix theory. Here we present a semiclassical derivation showing the validity of random matrix results. In order to simplify the semiclassical treatment, we express the moments of the delay times in terms of correlation functions of scattering matrices at different energies. In the semiclassical approximation, the elements of the scattering matrix are given in terms of the classical scattering trajectories, requiring one to study correlations between sets of such trajectories. We describe the structure of correlated sets of trajectories and formulate the rules for their evaluation to the leading order in inverse channel number. This allows us to derive a polynomial equation satisfied by the generating function of the moments. Along with showing the agreement of our semiclassical results with the moments predicted by random matrix theory, we infer that the scattering matrix is unitary to all orders in the semiclassical approximation. △ Less

Submitted 7 January, 2010; v1 submitted 1 October, 2009; originally announced October 2009.

Comments: Refereed version. 18 pages, 5 figures

Journal ref: J Phys A 43 (2010) 035101

Semiclassical gaps in the density of states of chaotic Andreev billiards

Authors: Jack Kuipers, Daniel Waltner, Cyril Petitjean, Gregory Berkolaiko, Klaus Richter

Abstract: The connection of a superconductor to a chaotic ballistic quantum dot leads to interesting phenomena, most notably the appearance of a hard gap in its excitation spectrum. Here we treat such an Andreev billiard semiclassically where the density of states is expressed in terms of the classical trajectories of electrons (and holes) that leave and return to the superconductor. We show how classical… ▽ More The connection of a superconductor to a chaotic ballistic quantum dot leads to interesting phenomena, most notably the appearance of a hard gap in its excitation spectrum. Here we treat such an Andreev billiard semiclassically where the density of states is expressed in terms of the classical trajectories of electrons (and holes) that leave and return to the superconductor. We show how classical orbit correlations lead to the formation of the hard gap, as predicted by random matrix theory in the limit of negligible Ehrenfest time $\tE$, and how the influence of a finite $\tE$ causes the gap to shrink. Furthermore, for intermediate $\tE$ we predict a second gap below $E=π\hbar /2\tE$ which would presumably be the clearest signature yet of $\tE$-effects. △ Less

Submitted 14 January, 2010; v1 submitted 15 July, 2009; originally announced July 2009.

Comments: Refereed version. 4 pages, 3 figures

Journal ref: Phys Rev Lett 104 (2010) 027001

Semiclassics for chaotic systems with tunnel barriers

Authors: Jack Kuipers

Abstract: The addition of tunnel barriers to open chaotic systems, as well as representing more general physical systems, leads to much richer semiclassical dynamics. In particular, we present here a complete semiclassical treatment for these systems, in the regime where Ehrenfest time effects are negligible and for times shorter than the Heisenberg time. To start we explore the trajectory structures whic… ▽ More The addition of tunnel barriers to open chaotic systems, as well as representing more general physical systems, leads to much richer semiclassical dynamics. In particular, we present here a complete semiclassical treatment for these systems, in the regime where Ehrenfest time effects are negligible and for times shorter than the Heisenberg time. To start we explore the trajectory structures which contribute to the survival probability, and find results that are also in agreement with random matrix theory. Then we progress to the treatment of the probability current density and are able to show, using recursion relation arguments, that the continuity equation connecting the current density to the survival probability is satisfied to all orders in the semiclassical approximation. Following on, we also consider a correlation function of the scattering matrix, for which we have to treat a new set of possible trajectory diagrams. By simplifying the contributions of these diagrams, we show that the results obtained here are consistent with known properties of the scattering matrix. The correlation function can be trivially connected to the ac and dc conductances, quantities of particular interest for which finally we present a semiclassical expansion. △ Less

Submitted 21 September, 2009; v1 submitted 10 June, 2009; originally announced June 2009.

Comments: Refereed version. 26 pages, 3 figures in 6 parts

Journal ref: J Phys A 42 (2009) 425101

Semiclassical approach to the ac-conductance of chaotic cavities

Authors: Cyril Petitjean, Daniel Waltner, Jack Kuipers, Inanc Adagideli, Klaus Richter

Abstract: We address frequency-dependent quantum transport through mesoscopic conductors in the semiclassical limit. By generalizing the trajectory-based semiclassical theory of dc quantum transport to the ac case, we derive the average screened conductance as well as ac weak-localization corrections for chaotic conductors. Thereby we confirm respective random matrix results and generalize them by account… ▽ More We address frequency-dependent quantum transport through mesoscopic conductors in the semiclassical limit. By generalizing the trajectory-based semiclassical theory of dc quantum transport to the ac case, we derive the average screened conductance as well as ac weak-localization corrections for chaotic conductors. Thereby we confirm respective random matrix results and generalize them by accounting for Ehrenfest time effects. We consider the case of a cavity connected through many leads to a macroscopic circuit which contains ac-sources. In addition to the reservoir the cavity itself is capacitively coupled to a gate. By incorporating tunnel barriers between cavity and leads we obtain results for arbitrary tunnel rates. Finally, based on our findings we investigate the effect of dephasing on the charge relaxation resistance of a mesoscopic capacitor in the linear low-frequency regime. △ Less

Submitted 9 June, 2009; originally announced June 2009.

Journal ref: Phys. Rev. B, 80,115310 (2009)

Non-Markovian dynamics of clusters during nucleation

Authors: J. Kuipers, G. T. Barkema

Abstract: Most theories of homogeneous nucleation are based on a Fokker-Planck-like description of the behavior of the mass of clusters. Here we will show that these approaches are incomplete for a large class of nucleating systems, as they assume the effective dynamics of the clusters to be Markovian, i.e., memoryless. We characterize these non-Markovian dynamics and show how this influences the dynamics… ▽ More Most theories of homogeneous nucleation are based on a Fokker-Planck-like description of the behavior of the mass of clusters. Here we will show that these approaches are incomplete for a large class of nucleating systems, as they assume the effective dynamics of the clusters to be Markovian, i.e., memoryless. We characterize these non-Markovian dynamics and show how this influences the dynamics of clusters during nucleation. Our results are validated by simulations of a three-dimensional Ising model with locally conserved magnetization. △ Less

Submitted 11 May, 2009; originally announced May 2009.

Comments: 4 pages, 4 figures

The semiclassical continuity equation for open chaotic systems

Authors: Jack Kuipers, Daniel Waltner, Martha Gutierrez, Klaus Richter

Abstract: We consider the continuity equation for open chaotic quantum systems in the semiclassical limit. First we explicitly calculate a semiclassical expansion for the probability current density using an expression based on classical trajectories. The current density is related to the survival probability via the continuity equation, and we show that this relation is satisfied within the semiclassical… ▽ More We consider the continuity equation for open chaotic quantum systems in the semiclassical limit. First we explicitly calculate a semiclassical expansion for the probability current density using an expression based on classical trajectories. The current density is related to the survival probability via the continuity equation, and we show that this relation is satisfied within the semiclassical approximation to all orders. For this we develop recursion relation arguments which connect the trajectory structures involved for the survival probability, which travel from one point in the bulk to another, to those structures involved for the current density, which travel from the bulk to the lead. The current density can also be linked, via another continuity equation, to a correlation function of the scattering matrix whose semiclassical approximation is expressed in terms of trajectories that start and end in the lead. We also show that this continuity equation holds to all orders. △ Less

Submitted 22 June, 2009; v1 submitted 13 November, 2008; originally announced November 2008.

Comments: Refereed version: Minor changes in presentation (especially in the Introduction) and minor corrections to the start of section 6. 21 pages, 1 figure in two parts

Journal ref: Nonlinearity 22 (2009) 1945-1964