Source author record

Ryan Turner

Ryan Turner appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Machine Learning astro-ph.CO Artificial Intelligence astro-ph.GA Robotics Systems and Control

Catalog footprint

What is connected

5works

6topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2023arXiv

Testing modified gravity scenarios with direct peculiar velocities

The theoretical basis of dark energy remains unknown and could signify a need to modify the laws of gravity on cosmological scales. In this study we investigate how the clustering and motions of galaxies can be used as probes of modified gravity theories, using galaxy and direct peculiar velocity auto- and cross-correlation functions. We measure and fit these correlation functions in simulations of $Λ$CDM, DGP, and $f(R)$ cosmologies and, by extracting the characteristic parameters of each model, we show that these theories can be distinguished from General Relativity using these measurements. We present forecasts showing that with sufficiently large data samples, this analysis technique is a competitive probe that can help place limits on allowed deviations from GR. For example, a peculiar velocity survey reaching to $z=0.5$ with $20\%$ distance accuracy would constrain model parameters to 3-$σ$ confidence limits $\log_{10}|f_{R0}| < -6.45$ for $f(R)$ gravity and $r_c > 2.88 \, c/H_0$ for nDGP, assuming a fiducial GR model.

preprint2022arXiv

Galaxy And Mass Assembly (GAMA): Data Release 4 and the z < 0.1 total and z < 0.08 morphological galaxy stellar mass functions

In Galaxy And Mass Assembly Data Release 4 (GAMA DR4), we make available our full spectroscopic redshift sample. This includes 248682 galaxy spectra, and, in combination with earlier surveys, results in 330542 redshifts across five sky regions covering ~250deg^2. The redshift density, is the highest available over such a sustained area, has exceptionally high completeness (95 per cent to r_KIDS=19.65mag), and is well suited for the study of galaxy mergers, galaxy groups, and the low redshift (z<0.25) galaxy population. DR4 includes 32 value-added tables or Data Management Units (DMUs) that provide a number of measured and derived data products including GALEX, ESO KiDS, ESO VIKING, WISE and Herschel Space Observatory imaging. Within this release, we provide visual morphologies for 15330 galaxies to z<0.08, photometric redshift estimates for all 18million objects to r_KIDS~25mag, and stellar velocity dispersions for 111830 galaxies. We conclude by deriving the total galaxy stellar mass function (GSMF) and its sub-division by morphological class (elliptical, compact-bulge and disc, diffuse-bulge and disc, and disc only). This extends our previous measurement of the total GSMF down to 10^6.75 M_sol h^-2_70 and we find a total stellar mass density of rho_*=(2.97+/-0.04)x10^8 M_sol h_70 Mpc^-3 or Omega_*=(2.17+/-0.03)x10^-3 h^-1_70. We conclude that at z<0.1, the Universe has converted 4.9+/-0.1 per cent of the baryonic mass implied by Big Bang Nucleosynthesis into stars that are gravitationally bound within the galaxy population.

preprint2020arXiv

Scalable Global Optimization via Local Bayesian Optimization

Bayesian optimization has recently emerged as a popular method for the sample-efficient optimization of expensive black-box functions. However, the application to high-dimensional problems with several thousand observations remains challenging, and on difficult problems Bayesian optimization is often not competitive with other paradigms. In this paper we take the view that this is due to the implicit homogeneity of the global probabilistic models and an overemphasized exploration that results from global acquisition. This motivates the design of a local probabilistic approach for global optimization of large-scale high-dimensional problems. We propose the $\texttt{TuRBO}$ algorithm that fits a collection of local models and performs a principled global allocation of samples across these models via an implicit bandit approach. A comprehensive evaluation demonstrates that $\texttt{TuRBO}$ outperforms state-of-the-art methods from machine learning and operations research on problems spanning reinforcement learning, robotics, and the natural sciences.

preprint2016arXiv

A Model Explanation System: Latest Updates and Extensions

We propose a general model explanation system (MES) for "explaining" the output of black box classifiers. This paper describes extensions to Turner (2015), which is referred to frequently in the text. We use the motivating example of a classifier trained to detect fraud in a credit card transaction history. The key aspect is that we provide explanations applicable to a single prediction, rather than provide an interpretable set of parameters. We focus on explaining positive predictions (alerts). However, the presented methodology is symmetrically applicable to negative predictions.

preprint2012arXiv

Robust Filtering and Smoothing with Gaussian Processes

We propose a principled algorithm for robust Bayesian filtering and smoothing in nonlinear stochastic dynamic systems when both the transition function and the measurement function are described by non-parametric Gaussian process (GP) models. GPs are gaining increasing importance in signal processing, machine learning, robotics, and control for representing unknown system functions by posterior probability distributions. This modern way of "system identification" is more robust than finding point estimates of a parametric function representation. In this article, we present a principled algorithm for robust analytic smoothing in GP dynamic systems, which are increasingly used in robotics and control. Our numerical evaluations demonstrate the robustness of the proposed approach in situations where other state-of-the-art Gaussian filters and smoothers can fail.