We elaborate on the functional integral describing the stochastic dynamics of a spectator field during inflation, comparing its diagrammatic expansion to that obtained directly from a perturbative solution of the corresponding Langevin equation. We state Feynman rules for computing arbitrary temporal $n$-point functions and perform some illustrative computations for a $\lambda\phi^4$ interaction, paying attention to the role played by a functional Jacobian determinant in the path integral. We also briefly consider the case when the field contributes to the expansion rate, making the noise multiplicative, which introduces additional vertices.

We examine scalar quantum electrodynamics in power-law inflation space-time, and compute the one-loop correction to electric and magnetic field correlators at superhorizon separations. The effect at one-loop descends from the coupling of the vector to the charged scalar current which is greatly enhanced due to gravitational particle production. We conclude that non-perturbative effects must exist due to (i) secular growth, (ii) spatial running, and (iii) infrared sensitivity of the one-loop correction to the correlators. Electric and magnetic correlators exhibit a hierarchy that is due to Faraday's law and accelerated expansion, and must hold non-perturbatively.

We apply the functional renormalization group to Starobinsky's stochastic equation describing the local dynamics of a light scalar field in de Sitter. After elaborating on the over-damped regime of stochastic dynamics, we introduce an effective average action for the stochastic field, resulting by progressively integrating out frequencies, and study its flow equation in the local potential approximation (LPA). This effective action determines the approach to equilibrium and allows for the computation of unequal time correlators $\left\langle\phi(t)\phi(t+\Delta t)\right\rangle$ for large values of $\Delta t$. The stochastic RG flow in the LPA can be formulated in two ways, one that preserves the stochastic supersymmetry and one that breaks it. We show that both predict a characteristic decay time very close to that determined by the dynamical mass for a massless self-interacting scalar in de Sitter $m^2\sim \sqrt{\lambda}H^2$. Furthermore, the temporal supersymmetric formulation remarkably recovers the flow for the effective potential found using Quantum Field Theory methods and a smoothing over spatial wavelengths. We also discuss how the stochastic framework generically predicts an infrared mass which is a few percent smaller than the dynamical mass obtained in the LPA. Our results further support the notion that stochastic inflation captures the correct IR dynamics of light scalar fields in inflation.

We discuss the effective infrared theory governing a light scalar's long wavelength dynamics in de Sitter spacetime. We show how the separation of scales around the physical curvature radius $k/a \sim H$ can be performed consistently with a window function and how short wavelengths can be integrated out in the Schwinger-Keldysh path integral formalism. At leading order, and for time scales $\Delta t \gg H^{-1}$, this results in the well-known Starobinsky stochastic evolution. However, our approach allows for the computation of quantum UV corrections, generating an effective potential on which the stochastic dynamics takes place. The long wavelength stochastic dynamical equations are now second order in time, incorporating temporal scales $\Delta t \sim H^{-1}$ and resulting in a Kramers equation for the probability distribution - more precisely the Wigner function - in contrast to the more usual Fokker-Planck equation. This feature allows us to non-perturbatively evaluate, within the stochastic formalism, not only expectation values of field correlators, but also the stress-energy tensor of $\phi$.

The infrared dynamics of a light, minimally coupled scalar field in de Sitter spacetime with Ricci curvature $R=12H$, averaged over horizon sized regions of physical volume $V_H=\frac{4\pi}{3}\left(\frac{1}{H}\right)^3$, can be interpreted as Brownian motion in a medium with de Sitter temperature $T_{DS}=\frac{\hbar H}{2\pi}$. We demonstrate this by directly deriving the effective action of scalar field fluctuations with wavelengths larger than the de Sitter curvature radius and generalizing Starobinsky's seminal results on stochastic inflation. The effective action describes stochastic dynamics and the fluctuating force drives the field to an equilibrium characterized by a thermal Gibbs distribution at temperature $T_{DS}$ which corresponds to a de Sitter invariant state. Hence, approach towards this state can be interpreted as thermalization. We show that the stochastic kinetic energy of the coarse-grained description corresponds to the norm of $\partial_\mu\phi$ and takes a well defined value per horizon volume $\frac{1}{2}\langle \left(\nabla\phi\right)^2\rangle = - \frac{1}{2}T_{DS}/V_H$. This approach allows for the non-perturbative computation of the de Sitter invariant stress energy tensor $\langle T_{\mu\nu} \rangle$ for an arbitrary scalar potential.

Using the Stochastic Adhesion Model (SAM) as a simple toy model for cosmic structure formation, we study renormalization and the removal of the cutoff dependence from loop integrals in perturbative calculations. SAM shares the same symmetry with the full system of continuity+Euler equations and includes a viscosity term and a stochastic noise term, similar to the effective theories recently put forward to model CDM clustering. We show in this context that if the viscosity and noise terms are treated as perturbative corrections to the standard eulerian perturbation theory, they are necessarily non-local in time. To ensure Galilean Invariance higher order vertices related to the viscosity and the noise must then be added and we explicitly show at one-loop that these terms act as counter terms for vertex diagrams. The Ward Identities ensure that the non-local-in-time theory can be renormalized consistently. Another possibility is to include the viscosity in the linear propagator, resulting in exponential damping at high wavenumber. The resulting local-in-time theory is then renormalizable to one loop, requiring less free parameters for its renormalization.

We consider a massive scalar field with quartic self-interaction $\lambda/4!\,\phi^4$ in de~Sitter spacetime and present a diagrammatic expansion that describes the field as driven by stochastic noise. This is compared with the Feynman diagrams in the Keldysh basis of the Amphichronous (Closed-Time-Path) Field Theoretical formalism. For all orders in the expansion, we find that the diagrams agree when evaluated in the leading infrared approximation, i.e. to leading order in $m^2/H^2$, where $m$ is the mass of the scalar field and $H$ is the Hubble rate. As a consequence, the correlation functions computed in both approaches also agree to leading infrared order. This perturbative correspondence shows that the stochastic Theory is exactly equivalent to the Field Theory in the infrared. The former can then offer a non-perturbative resummation of the Field Theoretical Feynman diagram expansion, including fields with $0\leq m^2\ll\sqrt \lambda H^2$ for which the perturbation expansion fails at late times.

The adhesion model has been proposed in the past as an improvement of the Zel'dovich approximation, providing a good description of the formation of the cosmic web. We recast the model as a field theory for cosmological large scale structure, adding a stochastic force to account for power generated from very short, highly non-linear scales that is uncorrelated with the initial power spectrum. The dynamics of this Stochastic Adhesion Model (SAM) is reminiscent of the well known Kardar-Parisi-Zhang equation with the difference that the viscosity and the noise spectrum are time dependent. Choosing the viscosity proportional to the growth factor $D$ restricts the form of noise spectrum through a 1-loop renormalization argument. For this choice, the SAM field theory is renormalizable to one loop. We comment on the suitability of this model for describing the non-linear regime of the CDM power spectrum and its utility as a relatively simple approach to cosmological clustering.

We discuss the relation between the output of Newtonian N-body simulations on scales that approach or exceed the particle horizon to the description of General Relativity. At leading order, the Zeldovich approximation is correct on large scales, coinciding with the General Relativistic result. At second order in the initial metric potential, the trajectories of particles deviate from the second order Newtonian result and hence the validity of 2LPT initial conditions should be reassessed when used in very large simulations. We also advocate using the expression for the synchronous gauge density as a well behaved measure of density fluctuations on such scales.

We examine the deviation of Cold Dark Matter particle trajectories from the Newtonian result as the size of the region under study becomes comparable to or exceeds the particle horizon. To first order in the gravitational potential, the general relativistic result coincides with the Zel'dovich approximation and hence the Newtonian prediction on all scales. At second order, General Relativity predicts corrections which overtake the corresponding second order Newtonian terms above a certain scale of the order of the Hubble radius. However, since second order corrections are very much suppressed on such scales, we conclude that simulations which exceed the particle horizon but use Newtonian equations to evolve the particles, reproduce the correct trajectories very well. The dominant relativistic corrections to the power spectrum on scales close to the horizon are at most of the order of $\sim 10^{-5}$ at $z=49$ and $\sim 10^{-3}$ at $z=0$. The differences in the positions of real space features are affected at a level below $10^{-6}$ at both redshifts. Our analysis also clarifies the relation of N-body results to relativistic considerations.

We describe the irrotational dust component of the universe in terms of a relativistic gradient expansion and transform the resulting synchronous metric to a Newtonian coordinate system. The two metrics are connected via a space-like displacement field and a time-like perturbation, providing a relativistic generalization of the transformation from Lagrangian to Eulerian coordinates. The relativistic part of the displacement field generates already at initial time a non-local density perturbation at second order. This is a purely relativistic effect since it originates from space-time mixing. We give two options, the passive and the active approach, on how to include the relativistic corrections for example in N-body simulations. In the passive approach we treat the corrections as a non-Gaussian modification of the initial Gaussian field (primordial non-Gaussianity could be incorporated as well). The induced non-Gaussianity depends on scale and the redshift at which initial conditions are set, with f_NL ~ few for small enough scales and redshifts. In the active approach we show how to use the relativistic trajectory to obtain the initial displacement and velocity of particles for N-body simulations without modifying the initial Gaussian field.

The infrared dynamics of a minimally coupled scalar field in de Sitter spacetime can be described as Brownian motion of a particle in a medium of de Sitter temperature $T_{DS}=\frac{H}{2\pi}$. The system obeys a fluctuation-dissipation relation and its equilibrium distribution is Maxwell-Boltzmann, implying kinetic and potential energies of comparable magnitudes. The transition to equilibrium is a semi-classical process beyond the scope of perturbation theory for interacting fields. The stochastic kinetic energy of the field causes de Sitter spacetime to cool down slowly with a corresponding decrease of the effective vacuum energy.

We show how the Zel'dovich approximation and the second order displacement field of Lagrangian perturbation theory can be obtained from a general relativistic gradient expansion in \Lambda{}CDM cosmology. The displacement field arises as a result of a second order non-local coordinate transformation which brings the synchronous/comoving metric into a Newtonian form. We find that, with a small modification, the Zel'dovich approximation holds even on scales comparable to the horizon. The corresponding density perturbation is not related to the Newtonian potential via the usual Poisson equation but via a modified Helmholtz equation. This is a consequence of causality not present in the Newtonian theory. The second order displacement field receives relativistic corrections that are subdominant on short scales but are comparable to the second order Newtonian result on scales approaching the horizon. The corrections are easy to include when setting up initial conditions in large N-body simulations.

We describe inhomogeneities in a {\Lambda}CDM universe with a gradient series expansion and show that it describes the gravitational evolution far into the non-linear regime and beyond the capacity of standard perturbation theory at any order. We compare the gradient expansion with exact inhomogeneous {\Lambda}LTB solutions (Lema\^itre-Tolman-Bondi metric with the inclusion of a cosmological constant) describing growing structure in a {\Lambda}CDM universe and find that the expansion approximates the exact solution well, following the collapse of an over-density all the way into a singularity.

We address the issue of cosmological backreaction from non-linear structure formation by constructing an approximation for the time evolved metric of a dust dominated universe based on a gradient expansion. Our metric begins as a perturbation of a flat Friedmann-Robertson-Walker state described by a nearly scale invariant, Gaussian, power-law distribution, and evolves in time until non-linear structures have formed. After describing and attempting to control for certain complications in the implementation of this approach, this metric then forms a working model of the universe. We numerically calculate the evolution of the average scale factor in this model and hence the backreaction. We argue that, despite its limitations, this model is more realistic than previous models that have confronted the issue of backreaction. We find that the \emph{instantaneous} effects of backreaction in this model could be as large as $\sim10%$ of the background. This suggests that a proper understanding of the \emph{cumulative} effects of backreaction could be crucial for precision cosmology and any future exploration of the dark sector.

We consider a set of scalar fields, consisting of a single flat direction and one or several non-flat directions. We take our cue from the MSSM, considering separately D-flat and F-flat directions, but our results apply to any supersymmetric scenario containing flat directions. We study the field fluctuations during pure de Sitter Inflation, following the evolution of the infrared modes by numerically solving the appropriate Langevin equations. We demonstrate that for the Standard Model U(1), SU(2) or SU(3) gauge couplings, as well as for large enough Yukawa couplings, the fluctuations along the non-flat directions effectively block the fluctuations along the flat directions. The usual expected behaviour <{\phi}^2> \propto N, with N the number of efolds, may be strongly violated, depending on the coupling strengths. As a consequence, those cosmological considerations, which are derived assuming that during inflation flat directions fluctuate freely, should be revised.

We clarify the role of gauge invariance for the computation of quantum non-Gaussian correlators in inflation. A gauge invariant generating functional for n-point functions is given and the special status of the spatially flat gauge is pointed out. We also comment on the relation between gauge transformations, field redefinitions, the choice of $t={\rm const}$ hypersurfaces and the use of boundary terms in computations of non-Gaussianity.

We consider inflation within the context of what is arguably the simplest non-metric extension of Einstein gravity. There non-metricity is described by a single graviscalar field with a non-minimal kinetic coupling to the inflaton field $\Psi$, parameterized by a single parameter $\gamma$. We discuss the implications of non-metricity for chaotic inflation and find that it significantly alters the inflaton dynamics for field values $\Psi \gtrsim M_P/\gamma$, dramatically changing the qualitative behaviour in this regime. For potentials with a positive slope non-metricity imposes an upper bound on the possible number of e-folds. For chaotic inflation with a monomial potential, the spectral index and the tensor-to-scalar ratio receive small corrections dependent on the non-metricity parameter. We also argue that significant post-inflationary non-metricity may be generated.

Self-energies of a minimally coupled scalar field with quartic and trilinear interactions are calculated in a de Sitter background, using a position space propagator. For quartic interactions, we recover earlier results for the seagull diagram, namely that it contributes an effective mass for the scalar field at leading order in the infrared enhancement in a steady-state de Sitter background. We further show that the sunset diagram also contributes to this effective mass and argue that these two contributions are sufficient in order to determine a self-consistent dynamical mass. In addition, trilinear interactions also induce a dynamical mass for the scalar field which we calculate. Since an interacting scalar field in de Sitter acquires a dynamical mass through these loop corrections, the infrared divergences of the two-point correlator are naturally self-regulated.

The quantum theory of cosmological perturbations in single field inflation is formulated in terms of a path integral. Starting from a canonical formulation, we show how the free propagators can be obtained from the well known gauge-invariant quadratic action for scalar and tensor perturbations, and determine the interactions to arbitrary order. This approach does not require the explicit solution of the energy and momentum constraints, a novel feature which simplifies the determination of the interaction vertices. The constraints and the necessary imposition of gauge conditions is reflected in the appearance of various commuting and anti-commuting auxiliary fields in the action. These auxiliary fields are not propagating physical degrees of freedom but need to be included in internal lines and loops in a diagrammatic expansion. To illustrate the formalism we discuss the tree-level 3-point and 4-point functions of the inflaton perturbations, reproducing the results already obtained by the methods used in the current literature. Loop calculations are left for future work.

Cosmic structures at small non-linear scales $k>L\sim 0.2 h $ Mpc$^{-1}$ have an impact on the longer (quasi-)linear wavelengths with $k<L$ via non-linear UV-IR mode coupling. We evaluate this effect for a $\Lambda$CDM universe applying the effective fluid method of Baumann, Nicolis, Senatore and Zaldarriaga. For $k<L$ the $\Lambda$CDM growth function for the density contrast is found to receive a scale dependent correction and an effective anisotropic stress sources a shift between the two gravitational potentials, setting $\phi$ - $\psi \neq 0$. Since such a situation is generically considered as a signature of modified gravity and/or dark energy, these effects should be taken into account before any conclusions on the dark sector are drawn from the interpretation of future observations.

We present a systematic study of the amplitude of the primordial perturbation in curvaton models with self-interactions, treating both renormalizable and non-renormalizable interactions. In particular, we consider the possibility that the curvaton energy density is subdominant at the time of the curvaton decay. We find that large regions in the parameter space give rise to the observed amplitude of primordial perturbation even for non-renormalizable curvaton potentials, for which the curvaton energy density dilutes fast. At the time of its decay, the curvaton energy density may typically be subdominant by a relative factor of 10^-3 and still produce the observed perturbation. Field dynamics turns out to be highly non-trivial, and for non-renormalizable potentials and certain regions of the parameter space we observe a non-monotonous relation between the final curvature perturbation and the initial curvaton value. In those cases, the time evolution of the primordial perturbation also displays an oscillatory behaviour before the curvaton decay.

We show that a spherically symmetric perturbation of a dust dominated $\Omega=1$ FRW universe in the Newtonian gauge can lead to an apparent acceleration of standard candles and provide a fit to the magnitude-redshift relation inferred from the supernovae data, while the perturbation in the gravitational potential remains small at all scales. We also demonstrate that the supernovae data does not necessarily imply the presence of some additional non-perturbative contribution by showing that any Lemaitre-Tolman-Bondi model fitting the supernovae data (with appropriate initial conditions) will be equivalent to a perturbed FRW spacetime along the past light cone.

We argue that the curvaton decay takes place most naturally by way of a broad parametric resonance. The mechanism is analogous to resonant inflaton decay but does not require any tuning of the curvaton coupling strength to other scalar fields. For low scale inflation and a correspondingly low mass scale for the curvaton, we speculate on observable consequences including the possibility of stochastic gravitational waves

We discuss the infrared divergences that appear to plague cosmological perturbation theory. We show that within the stochastic framework they are regulated by eternal inflation so that the theory predicts finite fluctuations. Using the $\Delta N$ formalism to one loop, we demonstrate that the infrared modes can be absorbed into additive constants and the coefficients of the diagrammatic expansion for the connected parts of two and three-point functions of the curvature perturbation. As a result, the use of any infrared cutoff below the scale of eternal inflation is permitted, provided that the background fields are appropriately redefined. The natural choice for the infrared cutoff would of course be the present horizon; other choices manifest themselves in the running of the correlators. We also demonstrate that it is possible to define observables that are renormalization group invariant. As an example, we derive a non-perturbative, infrared finite and renormalization point independent relation between the two-point correlators of the curvature perturbation for the case of the free single field.

After simplifying and improving the non-Gaussian formalism we developed in previous work, we derive a quantitative expression for the three-point correlator (bispectrum) of the curvature perturbation in general multiple-field inflation models. Our result describes the evolution of non-Gaussianity on superhorizon scales caused by the nonlinear influence of isocurvature perturbations on the adiabatic perturbation during inflation. We then study a simple quadratic two-field potential and find that when slow roll breaks down and the field trajectory changes direction in field space, the non-Gaussianity can become large. However, for the simple models studied to date, the magnitude of this non-Gaussianity decays away after the isocurvature mode is converted into the adiabatic mode.

We discuss the possible role of isocurvature perturbations for the quantum decoherence of the curvature perturbation during inflation. We point out that if the inflaton trajectory in field space is curved, the adiabatic mode is generically coupled to the isocurvature mode and thus tracing out the latter can cause the curvature perturbation to decohere. We explicitly investigate this suggestion in a model of inflation with two decoupled massive fields and find that decoherence is effective for a wide range of mass ratios. In particular, we calculate the entanglement entropy for this model and show that it grows after horizon exit, providing a quantitative measure for decoherence.

We investigate non-Gaussianity in general multiple-field inflation using the formalism we developed in earlier papers. We use a perturbative expansion of the non-linear equations to calculate the three-point correlator of the curvature perturbation analytically. We derive a general expression that involves only a time integral over background and linear perturbation quantities. We work out this expression explicitly for the two-field slow-roll case, and find that non-Gaussianity can be orders of magnitude larger than in the single-field case. In particular, the bispectrum divided by the square of the power spectrum can easily be of O(1-10), depending on the model. Our result also shows the explicit momentum dependence of the bispectrum. This conclusion of large non-Gaussianity is confirmed in a semi-analytic slow-roll investigation of a simple quadratic two-field model.

We develop a non-linear framework for describing long-wavelength perturbations in multiple-field inflation. The basic variables describing inhomogeneities are defined in a non-perturbative manner, are invariant under changes of time slicing on large scales and include both matter and metric perturbations. They are combinations of spatial gradients generalising the gauge-invariant variables of linear theory. Dynamical equations are derived and supplemented with stochastic source terms which provide the long-wavelength initial conditions determined from short-wavelength modes. Solutions can be readily obtained via numerical simulations or analytic perturbative expansions. The latter are much simpler than the usual second-order perturbation theory. Applications are given in a companion paper.

We present a method by which cosmological perturbations can be quantitatively studied in single and multi-field inflationary models beyond linear perturbation theory. A non-linear generalization of the gauge-invariant Sasaki-Mukhanov variables is used in a long-wavelength approximation. These generalized variables remain invariant under time slicing changes on long wavelengths. The equations they obey are relatively simple and can be formulated for a number of time slicing choices. Initial conditions are set after horizon crossing and the subsequent evolution is fully non-linear. We briefly discuss how these methods can be implemented numerically in the study of non-Gaussian signatures from specific inflationary models.

We present a simple way to calculate non-Gaussianity in inflation using fully non-linear equations on long wavelengths with stochastic sources to take into account the short-wavelength quantum fluctuations. Our formalism includes both scalar metric and matter perturbations, combining them into variables which are invariant under changes of time slicing in the long-wavelength limit. We illustrate this method with a perturbative calculation in the single-field slow-roll case. We also introduce a convenient choice of variables to graphically present the full momentum dependence of the three-point correlator.

We present a method for the study of second-order superhorizon perturbations in multi field inflationary models with non trivial kinetic terms. We utilise a change of coordinates in field space to separate isocurvature and adiabatic perturbations generalizing previous results. We also construct second order gauge invariant variables related to them. It is found that with an arbitrary metric in field space the isocurvature perturbation sources the gravitational potential on long wavelengths even for ``straight'' trajectories. The potential decouples from the isocurvature perturbations if the background fields' trajectory is a geodesic in field space. Taking nonlinear effects into account shows that, in general, the two types of perturbations couple to each other. This is an outline of a possible procedure to study nonlinear and non-Gaussian effects during multifield inflation.

In this letter we review the separate universe approach for cosmological perturbations and point out that it is essentially the lowest order approximation to a gradient expansion. Using this approach, one can study the nonlinear evolution of inhomogeneous spacetimes and find the conditions under which the long wavlength curvature perturbation can vary with time. When there is one degree of freedom or a well-defined equation of state the nonlinear long wavelength curvature perturbation remains constant. With more degrees of freedom it can vary and this variation is determined by the non-adiabatic pressure perturbation, exactly as in linear theory. We identify combinations of spatial vectors characterizing the curvature perturbation which are invariant under a change of time hypersurfaces.