Simple lessons from complex learning: what a neural network model learns about cosmic structure formation.

We train a neural network model to predict the full phase space evolution of cosmological N-body simulations. Its success implies that the neural network model is accurately approximating the Green's function expansion that relates the initial conditions of the simulations to its outcome at later times in the deeply nonlinear regime. We test the accuracy of this approximation by assessing its performance on well-understood simple cases that have either known exact solutions or well-understood expansions. These scenarios include spherical configurations, isolated plane waves, and two interacting plane waves: initial conditions that are very different from the Gaussian random fields used for training. We find our model generalizes well to these well-understood scenarios, demonstrating that the networks have inferred general physical principles and learned the nonlinear mode couplings from the complex, random Gaussian training data. These tests also provide a useful diagnostic for finding the model's strengths and weaknesses, and identifying strategies for model improvement. We also test the model on initial conditions that contain only transverse modes, a family of modes that differ not only in their phases but also in their evolution from the longitudinal growing modes used in the training set. When the network encounters these initial conditions that are orthogonal to the training set, the model fails completely. In addition to these simple configurations, we evaluate the model's predictions for the density, displacement, and momentum power spectra with standard initial conditions for N-body simulations. We compare these summary statistics against N-body results and an approximate, fast simulation method called COLA (COmoving Lagrangian Acceleration). Our model achieves percent level accuracy at nonlinear scales of k ∼ 1 Mpc - 1 h , representing a significant improvement over COLA.

Augmenting astrophysical scaling relations with machine learning: Application to reducing the Sunyaev-Zeldovich flux-mass scatter.

Complex astrophysical systems often exhibit low-scatter relations between observable properties (e.g., luminosity, velocity dispersion, oscillation period). These scaling relations illuminate the underlying physics, and can provide observational tools for estimating masses and distances. Machine learning can provide a fast and systematic way to search for new scaling relations (or for simple extensions to existing relations) in abstract high-dimensional parameter spaces. We use a machine learning tool called symbolic regression (SR), which models patterns in a dataset in the form of analytic equations. We focus on the Sunyaev-Zeldovich flux-cluster mass relation (YSZ - M), the scatter in which affects inference of cosmological parameters from cluster abundance data. Using SR on the data from the IllustrisTNG hydrodynamical simulation, we find a new proxy for cluster mass which combines YSZ and concentration of ionized gas (cgas): M ∝ Yconc3/5 ≡ YSZ3/5(1 - A cgas). Yconc reduces the scatter in the predicted M by ∼20 - 30% for large clusters (M ≳ 1014 h-1 M⊙), as compared to using just YSZ. We show that the dependence on cgas is linked to cores of clusters exhibiting larger scatter than their outskirts. Finally, we test Yconc on clusters from CAMELS simulations and show that Yconc is robust against variations in cosmology, subgrid physics, and cosmic variance. Our results and methodology can be useful for accurate multiwavelength cluster mass estimation from upcoming CMB and X-ray surveys like ACT, SO, eROSITA and CMB-S4.

A Bayesian neural network predicts the dissolution of compact planetary systems.

We introduce a Bayesian neural network model that can accurately predict not only if, but also when a compact planetary system with three or more planets will go unstable. Our model, trained directly from short N-body time series of raw orbital elements, is more than two orders of magnitude more accurate at predicting instability times than analytical estimators, while also reducing the bias of existing machine learning algorithms by nearly a factor of three. Despite being trained on compact resonant and near-resonant three-planet configurations, the model demonstrates robust generalization to both nonresonant and higher multiplicity configurations, in the latter case outperforming models fit to that specific set of integrations. The model computes instability estimates up to [Formula: see text] times faster than a numerical integrator, and unlike previous efforts provides confidence intervals on its predictions. Our inference model is publicly available in the SPOCK (https://github.com/dtamayo/spock) package, with training code open sourced (https://github.com/MilesCranmer/bnn_chaos_model).

Using the Marked Power Spectrum to Detect the Signature of Neutrinos in Large-Scale Structure.

Cosmological neutrinos have their greatest influence in voids: These are the regions with the highest neutrino to dark matter density ratios. The marked power spectrum can be used to emphasize low-density regions over high-density regions and, therefore, is potentially much more sensitive than the power spectrum to the effects of neutrino masses. Using 22 000 N-body simulations from the Quijote suite, we quantify the information content in the marked power spectrum of the matter field and show that it outperforms the standard power spectrum by setting constraints improved by a factor larger than 2 on all cosmological parameters. The combination of marked and standard power spectra allows us to place a 4.3σ constraint on the minimum sum of the neutrino masses with a volume equal to 1 (Gpc h^{-1})^{3} and without cosmic microwave background priors. Combinations of different marked power spectra yield a 6σ constraint within the same conditions.

What Is the Price of Abandoning Dark Matter? Cosmological Constraints on Alternative Gravity Theories.

Any successful alternative gravity theory that obviates the need for dark matter must fit our cosmological observations. Measurements of microwave background polarization trace the large-scale baryon velocity field at recombination and show very strong O(1) baryon acoustic oscillations. Measurements of the large-scale structure of galaxies at low redshift show much weaker features in the spectrum. If the alternative gravity theory's dynamical equations for the growth rate of structure are linear, then the density field growth can be described by a Green's function: δ(x[over →],t)=δ(x[over →],t^{'})G(x,t,t^{'}). We show that the Green's function G(x,t,t^{'}) must have dramatic features that erase the initial baryon oscillations. This implies an acceleration law that changes sign on the ∼150 Mpc scale. On the other hand, if the alternative gravity theory has a large nonlinear term that couples modes on different scales, then the theory would predict large-scale non-Gaussian features in large-scale structure. These are not seen in the distribution of galaxies nor in the distribution of quasars. No proposed alternative gravity theory for dark matter seems to satisfy these constraints.

Predicting the long-term stability of compact multiplanet systems.

We combine analytical understanding of resonant dynamics in two-planet systems with machine-learning techniques to train a model capable of robustly classifying stability in compact multiplanet systems over long timescales of [Formula: see text] orbits. Our Stability of Planetary Orbital Configurations Klassifier (SPOCK) predicts stability using physically motivated summary statistics measured in integrations of the first [Formula: see text] orbits, thus achieving speed-ups of up to [Formula: see text] over full simulations. This computationally opens up the stability-constrained characterization of multiplanet systems. Our model, trained on ∼100,000 three-planet systems sampled at discrete resonances, generalizes both to a sample spanning a continuous period-ratio range, as well as to a large five-planet sample with qualitatively different configurations to our training dataset. Our approach significantly outperforms previous methods based on systems' angular momentum deficit, chaos indicators, and parametrized fits to numerical integrations. We use SPOCK to constrain the free eccentricities between the inner and outer pairs of planets in the Kepler-431 system of three approximately Earth-sized planets to both be below 0.05. Our stability analysis provides significantly stronger eccentricity constraints than currently achievable through either radial velocity or transit-duration measurements for small planets and within a factor of a few of systems that exhibit transit-timing variations (TTVs). Given that current exoplanet-detection strategies now rarely allow for strong TTV constraints [S. Hadden, T. Barclay, M. J. Payne, M. J. Holman, Astrophys. J. 158, 146 (2019)], SPOCK enables a powerful complementary method for precisely characterizing compact multiplanet systems. We publicly release SPOCK for community use.

Measurement of a Cosmographic Distance Ratio with Galaxy and Cosmic Microwave Background Lensing.

We measure the gravitational lensing shear signal around dark matter halos hosting constant mass galaxies using light sources at z∼1 (background galaxies) and at the surface of last scattering at z∼1100 (the cosmic microwave background). The galaxy shear measurement uses data from the CFHTLenS survey, and the microwave background shear measurement uses data from the Planck satellite. The ratio of shears from these cross-correlations provides a purely geometric distance measurement across the longest possible cosmological lever arm. This is because the matter distribution around the halos, including uncertainties in galaxy bias and systematic errors such as miscentering, cancels in the ratio for halos in thin redshift slices. We measure this distance ratio in three different redshift slices of the constant mass (CMASS) sample and combine them to obtain a 17% measurement of the distance ratio, r=0.390_{-0.062}^{+0.070}, at an effective redshift of z=0.53. This is consistent with the predicted ratio from the Planck best-fit cold dark matter model with a cosmological constant cosmology of r=0.419.

Kinematic Sunyaev-Zel'dovich Effect with Projected Fields: A Novel Probe of the Baryon Distribution with Planck, WMAP, and WISE Data.

The kinematic Sunyaev-Zel'dovich (KSZ) effect-the Doppler boosting of cosmic microwave background (CMB) photons due to Compton scattering off free electrons with nonzero bulk velocity-probes the abundance and the distribution of baryons in the Universe. All KSZ measurements to date have explicitly required spectroscopic redshifts. Here, we implement a novel estimator for the KSZ-large-scale structure cross-correlation based on projected fields: it does not require redshift estimates for individual objects, allowing KSZ measurements from large-scale imaging surveys. We apply this estimator to cleaned CMB temperature maps constructed from Planck and WMAP data and a galaxy sample from the Wide-field Infrared Survey Explorer (WISE). We measure the KSZ effect at 3.8σ-4.5σ significance, depending on the use of additional WISE galaxy bias constraints. We verify that our measurements are robust to possible dust emission from the WISE galaxies. Assuming the standard Λ cold dark matter cosmology, we directly constrain (f_{b}/0.158)(f_{free}/1.0)=1.48±0.19 (statistical error only) at redshift z≈0.4, where f_{b} is the fraction of matter in baryonic form and f_{free} is the free electron fraction. This is the tightest KSZ-derived constraint reported to date on these parameters. Astronomers have long known that baryons do not trace dark matter on ∼ kiloparsec scales and there has been strong evidence that galaxies are baryon poor. The consistency between the f_{b} value found here and the values inferred from analyses of the primordial CMB and big bang nucleosynthesis verifies that baryons approximately trace the dark matter distribution down to ∼ megaparsec scales. While our projected-field estimator is already competitive with other KSZ approaches when applied to current data sets (because we are able to use the full-sky WISE photometric survey), it will yield enormous signal-to-noise ratios when applied to upcoming high-resolution, multifrequency CMB surveys.

Evidence of Halo Assembly Bias in Massive Clusters.

We present significant evidence of halo assembly bias for SDSS redMaPPer galaxy clusters in the redshift range [0.1, 0.33]. By dividing the 8,648 clusters into two subsamples based on the average member galaxy separation from the cluster center, we first show that the two subsamples have very similar halo mass of M_{200m}≃1.9×10^{14} h^{-1}M_{⊙} based on the weak lensing signals at small radii R≲10 h^{-1}Mpc. However, their halo bias inferred from both the large-scale weak lensing and the projected autocorrelation functions differs by a factor of ∼1.5, which is a signature of assembly bias. The same bias hypothesis for the two subsamples is excluded at 2.5σ in the weak lensing and 4.4σ in the autocorrelation data, respectively. This result could bring a significant impact on both galaxy evolution and precision cosmology.

Evidence of lensing of the cosmic microwave background by dark matter halos.

We present evidence of the gravitational lensing of the cosmic microwave background by 10(13) solar mass dark matter halos. Lensing convergence maps from the Atacama Cosmology Telescope Polarimeter (ACTPol) are stacked at the positions of around 12 000 optically selected CMASS galaxies from the SDSS-III/BOSS survey. The mean lensing signal is consistent with simulated dark matter halo profiles and is favored over a null signal at 3.2σ significance. This result demonstrates the potential of microwave background lensing to probe the dark matter distribution in galaxy group and galaxy cluster halos.

The dark side of cosmology: dark matter and dark energy.

A simple model with only six parameters (the age of the universe, the density of atoms, the density of matter, the amplitude of the initial fluctuations, the scale dependence of this amplitude, and the epoch of first star formation) fits all of our cosmological data . Although simple, this standard model is strange. The model implies that most of the matter in our Galaxy is in the form of "dark matter," a new type of particle not yet detected in the laboratory, and most of the energy in the universe is in the form of "dark energy," energy associated with empty space. Both dark matter and dark energy require extensions to our current understanding of particle physics or point toward a breakdown of general relativity on cosmological scales.

Taking the Universe's Temperature with Spectral Distortions of the Cosmic Microwave Background.

The cosmic microwave background (CMB) energy spectrum is a near-perfect blackbody. The standard model of cosmology predicts small spectral distortions to this form, but no such distortion of the sky-averaged CMB spectrum has yet been measured. We calculate the largest expected distortion, which arises from the inverse Compton scattering of CMB photons off hot, free electrons, known as the thermal Sunyaev-Zel'dovich (TSZ) effect. We show that the predicted signal is roughly one order of magnitude below the current bound from the COBE-FIRAS experiment, but it can be detected at enormous significance (≳1000σ) by the proposed Primordial Inflation Explorer (PIXIE). Although cosmic variance reduces the effective signal-to-noise ratio to 230σ, this measurement will still yield a subpercent constraint on the total thermal energy of electrons in the observable Universe. Furthermore, we show that PIXIE can detect subtle relativistic effects in the sky-averaged TSZ signal at 30σ, which directly probe moments of the optical depth-weighted intracluster medium electron temperature distribution. These effects break the degeneracy between the electron density and the temperature in the mean TSZ signal, allowing a direct inference of the mean baryon density at low redshift. Future spectral distortion probes will thus determine the global thermodynamic properties of ionized gas in the Universe with unprecedented precision. These measurements will impose a fundamental "integral constraint" on models of galaxy formation and the injection of feedback energy over cosmic time.

Evidence of galaxy cluster motions with the kinematic Sunyaev-Zel'dovich effect.

Using high-resolution microwave sky maps made by the Atacama Cosmology Telescope, we for the first time present strong evidence for motions of galaxy clusters and groups via microwave background temperature distortions due to the kinematic Sunyaev-Zel'dovich effect. Galaxy clusters are identified by their constituent luminous galaxies observed by the Baryon Oscillation Spectroscopic Survey, part of the Sloan Digital Sky Survey III. We measure the mean pairwise momentum of clusters, with a probability of the signal being due to random errors of 0.002, and the signal is consistent with the growth of cosmic structure in the standard model of cosmology.

Detection of the power spectrum of cosmic microwave background lensing by the Atacama Cosmology Telescope.

We report the first detection of the gravitational lensing of the cosmic microwave background through a measurement of the four-point correlation function in the temperature maps made by the Atacama Cosmology Telescope. We verify our detection by calculating the levels of potential contaminants and performing a number of null tests. The resulting convergence power spectrum at 2° angular scales measures the amplitude of matter density fluctuations on comoving length scales of around 100 Mpc at redshifts around 0.5 to 3. The measured amplitude of the signal agrees with Lambda cold dark matter cosmology predictions. Since the amplitude of the convergence power spectrum scales as the square of the amplitude of the density fluctuations, the 4σ detection of the lensing signal measures the amplitude of density fluctuations to 12%.

Evidence for dark energy from the cosmic microwave background alone using the Atacama Cosmology Telescope lensing measurements.

For the first time, measurements of the cosmic microwave background radiation (CMB) alone favor cosmologies with w = -1 dark energy over models without dark energy at a 3.2-sigma level. We demonstrate this by combining the CMB lensing deflection power spectrum from the Atacama Cosmology Telescope with temperature and polarization power spectra from the Wilkinson Microwave Anisotropy Probe. The lensing data break the geometric degeneracy of different cosmological models with similar CMB temperature power spectra. Our CMB-only measurement of the dark energy density Ω(Λ) confirms other measurements from supernovae, galaxy clusters, and baryon acoustic oscillations, and demonstrates the power of CMB lensing as a new cosmological tool.

Lenses in the forest: cross correlation of the Lyman-alpha flux with cosmic microwave background lensing.

We present a theoretical estimate for a new observable: the cross correlation between the Lyman-alpha flux fluctuations in quasar spectra and the convergence of the cosmic microwave background as measured along the same line of sight. As a first step toward the assessment of its detectability, we estimate the signal-to-noise ratio using linear theory. Although the signal-to-noise is small for a single line of sight and peaks at somewhat smaller redshifts than those probed by the Lyman-alpha forest, we estimate a total signal-to-noise of 9 for cross correlating quasar spectra of SDSS-III with Planck and 20 for cross correlating with a future polarization based cosmic microwave background experiment. The detection of this effect would be a direct measure of the neutral hydrogen-matter cross correlation and could provide important information on the growth of structures at large scales in a redshift range which is still poorly probed.

Lithopanspermia in star-forming clusters.

This paper considers the lithopanspermia hypothesis in star-forming groups and clusters, where the chances of biological material spreading from one solar system to another is greatly enhanced (relative to action in the field) because of the close proximity of the systems and lower relative velocities. These effects more than compensate for the reduced time spent in such crowded environments. This paper uses approximately 300,000 Monte Carlo scattering calculations to determine the cross sections for rocks to be captured by binaries and provides fitting formulae for other applications. We assess the odds of transfer as a function of the ejection speed v (eject) and number N(.) of members in the birth aggregate. The odds of any given ejected meteoroid being recaptured by another solar system are relatively low, about 1:10(3)-10(6) over the expected range of ejection speeds and cluster sizes. Because the number of ejected rocks (with mass m > 10 kg) per system can be large, N (R) approximately 10(16), virtually all solar systems are likely to share rocky ejecta with all of the other solar systems in their birth cluster. The number of ejected rocks that carry living microorganisms is much smaller and less certain, but we estimate that N (B) approximately 10(7) rocks can be ejected from a biologically active solar system. For typical birth environments, the capture of life-bearing rocks is expected to occur N (bio) asymptotically equal to 10-16,000 times (per cluster), depending on the ejection speeds. Only a small fraction (f (imp) approximately 10(4)) of the captured rocks impact the surfaces of terrestrial planets, so that N (lps) asymptotically equal to 10(3)-1.6 lithopanspermia events are expected per cluster (under favorable conditions). Finally, we discuss the question of internal versus external seeding of clusters and the possibility of Earth seeding young clusters over its biologically active lifetime.

Optimal one-dimensional apodizations and shaped pupils for planet finding coronagraphy.

The realization that direct imaging of extrasolar planets could be technologically feasible within the next decade or so has inspired a great deal of recent research into high-contrast imaging. We have contributed several design ideas, all of which can be described as shaped pupil coronagraphs. We offer a complete and unified survey of one-dimensional shaped pupil designs, some of which have been published in our previous papers. We also introduce a promising new design, which we call bar-code masks. With these masks we can achieve the required contrast with a fairly large discovery zone and throughput, but most importantly they are perhaps the easiest to manufacture and might therefore stand up best to refined analyses.

Constraining the topology of the universe.

The first year data from the Wilkinson Microwave Anisotropy Probe are used to place stringent constraints on the topology of the Universe. We search for pairs of circles on the sky with similar temperature patterns along each circle. We restrict the search to back-to-back circle pairs, and to nearly back-to-back circle pairs, as this covers the majority of the topologies that one might hope to detect in a nearly flat universe. We do not find any matched circles with radius greater than 25 degrees. For a wide class of models, the nondetection rules out the possibility that we live in a universe with topology scale smaller than 24 Gpc.