ArviZ a unified library for exploratory analysis of Bayesian models in Python

Kumar, Ravin; Carroll, Colin; Hartikainen, Ari; Martin, Osvaldo A.

doi:10.21105/joss.01143

Cited by 437 publications

(299 citation statements)

References 12 publications

Supporting

Mentioning

261

Contrasting

Unclassified

Order By: Relevance

“…Two chains of 2000 tuning steps and 2000 samples each were used. Plots of results were produced using Arviz (Kumar et al, 2019) and Matplotlib (Hunter, 2007). A region of practical equivalence (ROPE) of 2SDs was established as main criterion for deciding whether posterior distributions provided evidence for effects, where 90% high posterior density intervals (HPDs) which overlapped substantially with the ROPE (over 80%) were considered null.…”

Section: Resultsmentioning

confidence: 99%

Semantic and Prosodic Threat Processing in Trait Anxiety

Moreno

Tuomainen

Vinson

2020

Preprint

View full text Add to dashboard Cite

The present study attempts to identify how trait anxiety, measured as worry-level, affects the processing of threatening speech Two experiments using dichotic listening tasks were implemented; where participants had to identify sentences that convey threat through three different information channels: prosody-only, semantic-only and both semantic and prosody (congruent threat). We expected different ear advantages (left or right) depending on task demands, information type, and worry level. We used a full Bayesian approach for statistical modelling and analysis. Results indicate that when participants made delayed responses (Experiment 1), reaction times increased with worry level, but under time pressure (Experiment 2) worry level only affected reaction times for semantically neutral, prosodically threatening stimuli. We explain this effect by proposing a fourth stage, associated with goal oriented deliberation, for a three-phasic multistep model of emotional language processing. Higher levels of trait anxiety could induce verbal repetitive thinking (i.e. worry and/or rumination), which might prolong the mentioned deliberation stages, thus slowing-down responses.

show abstract

Section: Resultsmentioning

confidence: 99%

Semantic and Prosodic Threat Processing in Trait Anxiety

Moreno

Tuomainen

Vinson

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The highest density interval is defined such that all points within the interval have a higher probability density than all points outside the interval. We used the ArviZ python package(Kumar et al 2019) to compute it.…”

mentioning

confidence: 99%

Dynamical traceback age of theβPictoris moving group

Miret-Roig

Galli

Brandner

et al. 2020

A&A

100

View full text Add to dashboard Cite

Context. The β Pictoris moving group is one of the most well-known young associations in the solar neighbourhood and several members are known to host circumstellar discs, planets, and comets. Measuring its age precisely is essential to the study of several astrophysical processes, such as planet formation and disc evolution, which are strongly age-dependent. Aims. We aim to determine a precise and accurate dynamical traceback age for the β Pictoris moving group. Methods. Our sample combines the extremely precise Gaia DR2 astrometry with ground-based radial velocities measured in an homogeneous manner. We use an updated version of our algorithm to determine dynamical ages. The new approach takes into account a robust estimate of the spatial and kinematic covariance matrices of the association to improve the sample selection process and to perform the traceback analysis. Results. We estimate a dynamical age of 18.5−2.4+2.0 Myr for the β Pictoris moving group. We investigated the spatial substructure of the association at the time of birth and we propose the existence of a core of stars that is more concentrated. We also provide precise radial velocity measurements for 81 members of β Pic, including ten stars with the first determinations of their radial velocities. Conclusions. Our dynamical traceback age is three times more precise than previous traceback age estimates and, more importantly, for the first time it reconciles the traceback age with the most recent estimates of other dynamical, lithium depletion boundaries and isochronal ages. This has been possible thanks to the excellent astrometric and spectroscopic precisions, the homogeneity of our sample, and the detailed analysis of binaries and membership.

show abstract

“…There are visualization tools that seek to communicate inference results to users in a compact and relevant way, like ArviZ (Kumar et al, 2019), a unified library that provides tools for diagnostics and visualizations of Bayesian inference in Python. However, a complex Bayesian model could result in a high-dimensional posterior that would require unwieldy tables to present summary statistics or a multitude of visualizations that are difficult to grapple with.…”

Section: Introductionmentioning

confidence: 99%

Increasing Interpretability of Bayesian Probabilistic Programming Models Through Interactive Representations

Taka

Stein

Williamson

2020

Front. Comput. Sci.

View full text Add to dashboard Cite

Bayesian probabilistic modeling is supported by powerful computational tools like probabilistic programming and efficient Markov Chain Monte Carlo (MCMC) sampling. However, the results of Bayesian inference are challenging for users to interpret in tasks like decision-making under uncertainty or model refinement. Decision-makers need simultaneous insight into both the model's structure and its predictions, including uncertainty in inferred parameters. This enables better assessment of the risk overall possible outcomes compatible with observations and thus more informed decisions. To support this, we see a need for visualization tools that make probabilistic programs interpretable to reveal the interdependencies in probabilistic models and their inherent uncertainty. We propose the automatic transformation of Bayesian probabilistic models, expressed in a probabilistic programming language, into an interactive graphical representation of the model's structure at varying levels of granularity, with seamless integration of uncertainty visualization. This interactive graphical representation supports the exploration of the prior and posterior distribution of MCMC samples. The interpretability of Bayesian probabilistic programming models is enhanced through the interactive graphical representations, which provide human users with more informative, transparent, and explainable probabilistic models. We present a concrete implementation that translates probabilistic programs to interactive graphical representations and show illustrative examples for a variety of Bayesian probabilistic models.

show abstract

ArviZ a unified library for exploratory analysis of Bayesian models in Python

Abstract: License Authors of papers retain copyright and release the work under a Creative Commons Attribution 4.0 International License (CC-BY).

Cited by 437 publications

References 12 publications

Semantic and Prosodic Threat Processing in Trait Anxiety

Semantic and Prosodic Threat Processing in Trait Anxiety

Dynamical traceback age of theβPictoris moving group

Increasing Interpretability of Bayesian Probabilistic Programming Models Through Interactive Representations

Contact Info

Product

Resources

About