arXiv subjects

Daniel Turek

Publications and source records attributed to Daniel Turek.

10 recordsLinked to original sources

arXiv2025

Beyond Suboptimality: Resource-Rationality and Task Demands Shape the Complexity of Perceptual Representations

Early theories of perception as probabilistic inference propose that uncertainty about the interpretation of sensory input is represented as a probability distribution over many interpretations -- a relatively complex representation. However, critics argue that persistent demonstrations of suboptimal perceptual decision-making indicate limits in representational complexity. We contend that suboptimality arises not from genuine limits, but participants' resource-rational adaptations to task demands. For example, when tasks are solvable with minimal attention to stimuli, participants may neglect information needed for complex representations, relying instead on simpler ones that engender suboptimality. Across three experiments, we progressively reduced the efficacy of resource-rational strategies on a carefully controlled decision task. Model fits favored simple representations when resource-rational strategies were effective, and favored complex representations when ineffective, suggesting that perceptual representations can be simple or complex depending on task demands. We conclude that resource-rationality is an epistemic constraint for experimental design and essential to a complete theory of perception.

q-bio.NC↗

arXiv2021

When Rooks Miss: Probability through Chess

A famous (and hard) chess problem asks what is the maximum number of safe squares possible in placing $n$ queens on an $n\times n$ board. We examine related problems from placing $n$ rooks. We prove that as $n\to\infty$, the probability rapidly tends to 1 that the fraction of safe squares from a random placement converges to $1/e^2$. Our interest in the problem is showing how to view the involved algebra to obtain the simple, closed form limiting fraction. In particular, we see the power of many of the key concepts in probability: binary indicator variables, linearity of expectation, variances and covariances, Chebyshev's inequality, and Stirling's formula.

math.PR↗

arXiv2020

Bayesian Non-Parametric Detection Heterogeneity in Ecological Models

Detection heterogeneity is inherent to ecological data, arising from factors such as varied terrain or weather conditions, inconsistent sampling effort, or heterogeneity of individuals themselves. Incorporating additional covariates into a statistical model is one approach for addressing heterogeneity, but is no guarantee that any set of measurable covariates will adequately address the heterogeneity, and the presence of unmodelled heterogeneity has been shown to produce biases in the resulting inferences. Other approaches for addressing heterogeneity include the use of random effects, or finite mixtures of homogeneous subgroups. Here, we present a non-parametric approach for modelling detection heterogeneity for use in a Bayesian hierarchical framework. We employ a Dirichlet process mixture which allows a flexible number of population subgroups without the need to pre-specify this number of subgroups as in a finite mixture. We describe this non-parametric approach, then consider its use for modelling detection heterogeneity in two common ecological motifs: capture-recapture and occupancy modelling. For each, we consider a homogeneous model, finite mixture models, and the non-parametric approach. We compare these approaches using two simulation studies, and observe the non-parametric approach as the most reliable method for addressing varying degrees of heterogeneity. We also present two real-data examples, and compare the inferences resulting from each modelling approach. Analyses are carried out using the \texttt{nimble} package for \texttt{R}, which provides facilities for Bayesian non-parametric models.

stat.AP↗

arXiv2020

Bayesian inference for high-dimensional nonstationary Gaussian processes

In spite of the diverse literature on nonstationary spatial modeling and approximate Gaussian process (GP) methods, there are no general approaches for conducting fully Bayesian inference for moderately sized nonstationary spatial data sets on a personal laptop. For statisticians and data scientists who wish to learn about spatially-referenced data and conduct posterior inference and prediction with appropriate uncertainty quantification, the lack of such approaches and corresponding software is a significant limitation. In this paper, we develop methodology for implementing formal Bayesian inference for a general class of nonstationary GPs. Our novel approach uses pre-existing frameworks for characterizing nonstationarity in a new way that is applicable for small to moderately sized data sets via modern GP likelihood approximations. Posterior sampling is implemented using flexible MCMC methods, with nonstationary posterior prediction conducted as a post-processing step. We demonstrate our novel methods on two data sets, ranging from several hundred to several thousand locations, and compare our methodology with related statistical methods that provide off-the-shelf software. All of our methods are implemented in the freely available BayesNSGP software package for R.

stat.CO↗

arXiv2020

Sequential Monte Carlo Methods in the nimble R Package

nimble is an R package for constructing algorithms and conducting inference on hierarchical models. The nimble package provides a unique combination of flexible model specification and the ability to program model-generic algorithms. Specifically, the package allows users to code models in the BUGS language, and it allows users to write algorithms that can be applied to any appropriate model. In this paper, we introduce nimble's capabilities for state-space model analysis using sequential Monte Carlo (SMC) techniques. We first provide an overview of state-space models and commonly-used SMC algorithms. We then describe how to build a state-space model and conduct inference using existing SMC algorithms within nimble. SMC algorithms within nimble currently include the bootstrap filter, auxiliary particle filter, ensemble Kalman filter, IF2 method of iterated filtering, and a particle MCMC sampler. These algorithms can be run in R or compiled into C++ for more efficient execution. Examples of applying SMC algorithms to linear autoregressive models and a stochastic volatility model are provided. Finally, we give an overview of how model-generic algorithms are coded within nimble by providing code for a simple SMC algorithm. This illustrates how users can easily extend nimble's SMC methods in high-level code.

stat.CO↗

arXiv2019

Automatic adaptation of MCMC algorithms

Markov chain Monte Carlo (MCMC) methods are ubiquitous tools for simulation-based inference in many fields but designing and identifying good MCMC samplers is still an open question. This paper introduces a novel MCMC algorithm, namely, Auto Adapt MCMC. For sampling variables or blocks of variables, we use two levels of adaptation where the inner adaptation optimizes the MCMC performance within each sampler, while the outer adaptation explores the valid space of kernels to find the optimal samplers. We provide a theoretical foundation for our approach. To show the generality and usefulness of the approach, we describe a framework using only standard MCMC samplers as candidate samplers and some adaptation schemes for both inner and outer iterations. In several benchmark problems, we show that our proposed approach substantially outperforms other approaches, including an automatic blocking algorithm, in terms of MCMC efficiency and computational time.

stat.ME↗

arXiv2019

Diversity of Artists in Major U.S. Museums

The U.S. art museum sector is grappling with diversity. While previous work has investigated the demographic diversity of museum staffs and visitors, the diversity of artists in their collections has remained unreported. We conduct the first large-scale study of artist diversity in museums. By scraping the public online catalogs of 18 major U.S. museums, deploying a sample of 10,000 artist records comprising over 9,000 unique artists to crowdsourcing, and analyzing 45,000 responses, we infer artist genders, ethnicities, geographic origins, and birth decades. Our results are threefold. First, we provide estimates of gender and ethnic diversity at each museum, and overall, we find that 85% of artists are white and 87% are men. Second, we identify museums that are outliers, having significantly higher or lower representation of certain demographic groups than the rest of the pool. Third, we find that the relationship between museum collection mission and artist diversity is weak, suggesting that a museum wishing to increase diversity might do so without changing its emphases on specific time periods and regions. Our methodology can be used to broadly and efficiently assess diversity in other fields.

stat.AP↗

arXiv2016

Programming with models: writing statistical algorithms for general model structures with NIMBLE

We describe NIMBLE, a system for programming statistical algorithms for general model structures within R. NIMBLE is designed to meet three challenges: flexible model specification, a language for programming algorithms that can use different models, and a balance between high-level programmability and execution efficiency. For model specification, NIMBLE extends the BUGS language and creates model objects, which can manipulate variables, calculate log probability values, generate simulations, and query the relationships among variables. For algorithm programming, NIMBLE provides functions that operate with model objects using two stages of evaluation. The first stage allows specialization of a function to a particular model and/or nodes, such as creating a Metropolis-Hastings sampler for a particular block of nodes. The second stage allows repeated execution of computations using the results of the first stage. To achieve efficient second-stage computation, NIMBLE compiles models and functions via C++, using the Eigen library for linear algebra, and provides the user with an interface to compiled objects. The NIMBLE language represents a compilable domain-specific language (DSL) embedded within R. This paper provides an overview of the design and rationale for NIMBLE along with illustrative examples including importance sampling, Markov chain Monte Carlo (MCMC) and Monte Carlo expectation maximization (MCEM).

stat.CO↗

arXiv2016

Efficient Markov Chain Monte Carlo Sampling for Hierarchical Hidden Markov Models

Traditional Markov chain Monte Carlo (MCMC) sampling of hidden Markov models (HMMs) involves latent states underlying an imperfect observation process, and generates posterior samples for top-level parameters concurrently with nuisance latent variables. When potentially many HMMs are embedded within a hierarchical model, this can result in prohibitively long MCMC runtimes. We study combinations of existing methods, which are shown to vastly improve computational efficiency for these hierarchical models while maintaining the modeling flexibility provided by embedded HMMs. The methods include discrete filtering of the HMM likelihood to remove latent states, reduced data representations, and a novel procedure for dynamic block sampling of posterior dimensions. The first two methods have been used in isolation in existing application-specific software, but are not generally available for incorporation in arbitrary model structures. Using the NIMBLE package for R, we develop and test combined computational approaches using three examples from ecological capture-recapture, although our methods are generally applicable to any embedded discrete HMMs. These combinations provide several orders of magnitude improvement in MCMC sampling efficiency, defined as the rate of generating effectively independent posterior samples. In addition to being computationally significant for this class of hierarchical models, this result underscores the potential for vast improvements to MCMC sampling efficiency which can result from combinations of known algorithms.

stat.CO↗

arXiv2015

Automated Parameter Blocking for Efficient Markov-Chain Monte Carlo Sampling

Markov chain Monte Carlo (MCMC) sampling is an important and commonly used tool for the analysis of hierarchical models. Nevertheless, practitioners generally have two options for MCMC: utilize existing software that generates a black-box "one size fits all" algorithm, or the challenging (and time consuming) task of implementing a problem-specific MCMC algorithm. Either choice may result in inefficient sampling, and hence researchers have become accustomed to MCMC runtimes on the order of days (or longer) for large models. We propose an automated procedure to determine an efficient MCMC algorithm for a given model and computing platform. Our procedure dynamically determines blocks of parameters for joint sampling that result in efficient sampling of the entire model. We test this procedure using a diverse suite of example models, and observe non-trivial improvements in MCMC efficiency for many models. Our procedure is the first attempt at such, and may be generalized to a broader space of MCMC algorithms. Our results suggest that substantive improvements in MCMC efficiency may be practically realized using our automated blocking procedure, or variants thereof, which warrants additional study and application.

stat.CO↗