Charting galactic accelerations – II. How to ‘learn’ accelerations in the solar neighbourhood

Naik, Aneesh P.; An, J.; Burrage, Clare; Evans, N. W.

doi:10.1093/mnras/stac153

Cited by 9 publications

(3 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, we would like to point to a related, likewise promising method, developed by An et al ( 2021) and Naik et al (2022), which similarly builds on Green & Ting (2020). This closely related method similarly uses a normalizing flow to recover the gradients of the distribution function.…”

Section: Discussionmentioning

confidence: 99%

Deep Potential: Recovering the gravitational potential from a snapshot of phase space

Green¹,

Ting²,

Kamdar³

2022

Preprint

View full text Add to dashboard Cite

One of the major goals of the field of Milky Way dynamics is to recover the gravitational potential field. Mapping the potential would allow us to determine the spatial distribution of matter -both baryonic and dark -throughout the Galaxy. We present a novel method for determining the gravitational field from a snapshot of the phase-space positions of stars, based only on minimal physical assumptions, which makes use of recently developed tools from the field of deep learning. We first train a normalizing flow on a sample of observed six-dimensional phase-space coordinates of stars, obtaining a smooth, differentiable approximation of the distribution function. Using the Collisionless Boltzmann Equation, we then find the gravitational potential -represented by a feed-forward neural networkthat renders this distribution function stationary. This method, which we term "Deep Potential," is more flexible than previous parametric methods, which fit restricted classes of analytic models of the distribution function and potential to the data. We demonstrate Deep Potential on mock datasets, and demonstrate its robustness under various non-ideal conditions. Deep Potential is a promising approach to mapping the density of the Milky Way and other stellar systems, using rich datasets of stellar positions and kinematics now being provided by Gaia and ground-based spectroscopic surveys.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep Potential: Recovering the gravitational potential from a snapshot of phase space

Green¹,

Ting²,

Kamdar³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Our method does not require a model for the underlying galactic potential, nor does it require streams to delineate isoenergy curves in phase-space (i.e., stellar orbits). The output of our analysis is similar to that of Naik et al (2022), as we constrain galactic accelerations rather than a latent potential model. We assume that the potential is mostly static, or changing slowly.…”

Section: Introductionmentioning

confidence: 99%

Charting Galactic Accelerations with Stellar Streams and Machine Learning

Nibauer

Belokurov

Cranmer

et al. 2022

ApJ

View full text Add to dashboard Cite

We present a data-driven method for reconstructing the galactic acceleration field from phase-space (position and velocity) measurements of stellar streams. Our approach is based on a flexible and differentiable fit to the stream in phase-space, enabling a direct estimate of the acceleration vector along the stream. Reconstruction of the local acceleration field can be applied independently to each of several streams, allowing us to sample the acceleration field due to the underlying galactic potential across a range of scales. Our approach is methodologically different from previous works, as a model for the gravitational potential does not need to be adopted beforehand. Instead, our flexible neural-network-based model treats the stream as a collection of orbits with a locally similar mixture of energies, rather than assuming that the stream delineates a single stellar orbit. Accordingly, our approach allows for distinct regions of the stream to have different mean energies, as is the case for real stellar streams. Once the acceleration vector is sampled along the stream, standard analytic models for the galactic potential can then be rapidly constrained. We find our method recovers the correct parameters for a ground-truth triaxial logarithmic halo potential when applied to simulated stellar streams. Alternatively, we demonstrate that a flexible potential can be constrained with a neural network, and standard multipole expansions can also be constrained. Our approach is applicable to simple and complicated gravitational potentials alike and enables potential reconstruction from a fully data-driven standpoint using measurements of slowly phase-mixing tidal debris.

show abstract

“…Bulk processing of noisy, image data at radio and optical wavelengths to classify galaxies is well-advanced-see, for example, Clarke et al (2020), Canducci et al (2022) and Tang et al (2022). For single galaxies, orbit trajectory data and normalizing flows provide the capability to determine distribution functions and accelerations from the gravitational potential -see Green & Ting (2021), An et al (2021) and Naik et al (2022). An et al (2021) used a Hernquist (1990) model in their experiments.…”

Section: Introductionmentioning

confidence: 99%

Stellar Dynamical Modeling—Counting Conserved Quantities

Long

Mao

Wang

2023

Res. Astron. Astrophys.

View full text Add to dashboard Cite

Knowing the conserved quantities that a galaxy's stellar orbits conform to is important in helping us understand the stellar distribution and structures within the galaxy. Isolating integrals of motion and resonances are particularly important, non-isolating integrals less so. We compare the behavior and results of two methods for counting the number of conserved quantities, one based on the correlation integral approach and the other a more recent method using machine learning. Both methods use stellar orbit trajectories in phase space as their only input, and we create such trajectories from theoretical spherical, axisymmetric and triaxial model galaxies. The orbits have known isolating integrals and resonances. We find that neither method is fully effective in recovering the numbers of these quantities, nor in determining the number of non-isolating integrals. From a computer performance perspective, we find the correlation integral approach to be faster. Determining the algebraic formulae of (multiple) conserved quantities from the trajectories has not been possible due to the lack of an appropriate symbolic regression capability. Notwithstanding the shortcomings we have noted, it may that the methods are usable as part of a trajectory analysis tool kit.

show abstract

Charting galactic accelerations – II. How to ‘learn’ accelerations in the solar neighbourhood

Cited by 9 publications

References 49 publications

Deep Potential: Recovering the gravitational potential from a snapshot of phase space

Deep Potential: Recovering the gravitational potential from a snapshot of phase space

Charting Galactic Accelerations with Stellar Streams and Machine Learning

Stellar Dynamical Modeling—Counting Conserved Quantities

Contact Info

Product

Resources

About