The recently proposed theory of "Asymptotically Free Mimetic Gravity" is extended to the general non-homogeneous, spatially non-flat case. We present a modified theory of gravity which is free of higher derivatives of the metric. In this theory asymptotic freedom of gravity implies the existence of a minimal black hole with vanishing Hawking temperature. Introducing a spatial curvature dependent potential, we moreover obtain non-singular, bouncing modifications of spatially non-flat Friedmann and Bianchi universes.

We show that in asymptotically free mimetic gravity with limiting curvature the black hole singularity can be resolved and replaced by a static patch of de Sitter space. As a result of Hawking evaporation of these non-singular black holes, there remain stable remnants with vanishing Hawking temperature.

The mass of the graviton can be generated using a Brout-Englert-Higgs mechanism with four scalar fields. We show that when one of these fields is costrained as in mimetic gravity, the massive gravity obtained is ghost free and consistent. The mass term is not of the Fierz-Pauli type. There are only five degrees of freedom and the sixth degree of freedom associated with the Boulware-Deser ghost is constrained and replaced by mimetic matter to all orders. The van Dam-Veltman-Zakharov discontinuity is also absent.

We present a theory of ghost-free massive gravity where the mass of the graviton is generated through the Brout-Englert-Higgs (BEH) mechanism and one of the four scalar fields used is that of mimetic gravity. The mass term is not of the Fierz-Pauli type and the constraint eliminates the Boulware-Deser ghost which is absent to all orders. We perform a detailed analysis using the methods of cosmological perturbation theory and consider quantum fluctuations of the degrees of freedom of massive graviton and mimetic matter. It is shown that for three of the degrees of freedom of the graviton of mass $m$ the nonlinear corrections become comparable to the linear terms already at a length scale of order $m^{-\frac{1}{2}}$. Thus, at smaller scales they become strongly coupled and the graviton remains with two transverse degrees of freedom which get strongly coupled only at Planck scale. The mimetic field behaves as cold particles of half of the graviton mass and could well explain the source of dark matter in our universe. In the weakly coupled domain mimetic matter is completely decoupled from the massive graviton.

We consider the Schwarzschild black hole and show how, in a theory with limiting curvature, the physical singularity "inside it" is removed. The resulting spacetime is geodesically complete. The internal structure of this nonsingular black hole is analogus to Russian nesting dolls. Namely, after falling into the black hole of radius $r_{g}$, an observer, instead of being destroyed at the singularity, gets for a short time into the region with limiting curvature. After that he re-emerges in the near horizon region of a spacetime described by the Schwarzschild metric of a gravitational radius proportional to $r_{g}^{1/3}$. In the next cycle, after passing the limiting curvature, the observer finds himself within a black hole of even smaller radius proportional to $r_{g}^{1/9}$, and so on. Finally after few cycles he will end up in the spacetime where he remains forever at limiting curvature.

We find a simple modification of the longitudinal mode in General Relativity which incorporates the idea of limiting curvature. In this case the singularities in contracting Friedmann and Kasner universes are avoided, and instead, the universe has a regular bounce which takes place during the time inversely proportional to the square root of the limiting curvature. Away from the bounce, corrections to General Relativity are negligible. In addition the non-singluar modification of General Relativity delivers for free a realistic candidate for Dark Matter.

We study the validity of the Newtonian description of cosmological perturbations using the Lemaitre model, an exact spherically symmetric solution of Einstein's equation. This problem has been investigated in the past for the case of a dust fluid. Here, we extend the previous analysis to the more general case of a fluid with non-negligible pressure, and, for the numerical examples, we consider the case of radiation (P=\rho/3). We find that, even when the density contrast has a nonlinear amplitude, the Newtonian description of the cosmological perturbations using the gravitational potential \psi and the curvature potential \phi is valid as long as we consider sub-horizon inhomogeneities. However, the relation \psi+\phi={\cal O}(\phi^2), which holds for the case of a dust fluid, is not valid for a relativistic fluid and effective anisotropic stress is generated. This demonstrates the usefulness of the Lemaitre model which allows us to study in an exact nonlinear fashion the onset of anisotropic stress in fluids with non-negligible pressure. We show that this happens when the characteristic scale of the inhomogeneity is smaller than the sound horizon and that the deviation is caused by the nonlinear effect of the fluid's fast motion. We also find that \psi+\phi= \max[{\cal O}(\phi^2),{\cal O}(c_s^2\phi \, \delta)] for an inhomogeneity with density contrast \delta whose characteristic scale is smaller than the sound horizon, unless w is close to -1, where w and c_s are the equation of state parameter and the sound speed of the fluid, respectively. On the other hand, we expect \psi+\phi={\cal O}(\phi^2) to hold for an inhomogeneity whose characteristic scale is larger than the sound horizon, unless the amplitude of the inhomogeneity is large and w is close to -1.

We modify Einstein General Relativity by adding non-dynamical scalar fields to account simultaneously for both dark matter and dark energy. The dark energy in this case can be distributed in-homogeneously even within horizon scales. Its inhomogeneities can contribute to the late time integrated Sachs-Wolfe effect, possibly removing some of the low multipole anomalies in the temperature fluctuations of the CMB spectrum. The presence of the inhomogeneous dark matter also influences structure formation in the universe.

We find a rather unique extension of inflationary scenario which avoids selfreproduction and thus resolves the problems of multiverse, predictability and initial conditions. In this theory the amplitude of the cosmological perturbations is expressed entirely in terms of the total duration of inflation.

Fluctuations in the intensity and polarization of the cosmic microwave background (CMB) and the large-scale distribution of matter in the universe each contain clues about the nature of the earliest moments of time. The next generation of CMB and large-scale structure (LSS) experiments are poised to test the leading paradigm for these earliest moments---the theory of cosmic inflation---and to detect the imprints of the inflationary epoch, thereby dramatically increasing our understanding of fundamental physics and the early universe. A future CMB experiment with sufficient angular resolution and frequency coverage that surveys at least 1% of the sky to a depth of 1 uK-arcmin can deliver a constraint on the tensor-to-scalar ratio that will either result in a 5-sigma measurement of the energy scale of inflation or rule out all large-field inflation models, even in the presence of foregrounds and the gravitational lensing B-mode signal. LSS experiments, particularly spectroscopic surveys such as the Dark Energy Spectroscopic Instrument, will complement the CMB effort by improving current constraints on running of the spectral index by up to a factor of four, improving constraints on curvature by a factor of ten, and providing non-Gaussianity constraints that are competitive with the current CMB bounds.

We consider minimal extensions of the recently proposed Mimetic Dark Matter and show that by introducing a potential for the mimetic non-dynamical scalar field we can mimic nearly any gravitational properties of the normal matter. In particular, the mimetic matter can provide us with inflaton, quintessence and even can lead to a bouncing nonsingular universe. We also investigate the behaviour of cosmological perturbations due to a mimetic matter. We demonstrate that simple mimetic inflation can produce red-tilted scalar perturbations which are largely enhanced over gravity waves.

We reformulate Einstein's theory of gravity, isolating the conformal degree of freedom in a covariant way. This is done by introducing a physical metric defined in terms of an auxiliary metric and a scalar field appearing through its first derivatives. The resulting equations of motion split into a traceless equation obtained through variation with respect to the auxiliary metric and an additional differential equation for the trace part. As a result the conformal degree of freedom becomes dynamical even in the absence of matter. We show that this extra degree of freedom can mimic cold dark matter.

I consider the generic model independent predictions of the theory of quantum cosmological perturbations. To describe the stage of cosmic inflation, where these perturbations are amplified, the hydrodynamical approch is used. The inflationary stage is completely characterized by the deviation of the equation of state from cosmological constant which is a smooth function of the number of e-folds until the end of inflation. It is shown that in this case the spectral index should deviate from the flat one at least by 3 percent irrespective of any particular scenario. Given the value of the spectral index the lower bound on the amount of the gravitational waves produced is derived. Finally the relation between effective hydrodynamical description of inflation and inflationary scenarios is discussed.

Recent progress in observational cosmology, and especially the forthcoming PLANCK mission data, open new directions in so-called precision cosmology. In this paper we illustrate this statement considering the accuracy of cosmological determination of the two-quanta decay rate of 2s hydrogen atom state. We show that the PLANCK data will allow us to measure this decay rate significantly better than in the laboratory experiments.

We discuss observational consequences of the curvaton scenario, which naturally appears in the context of the simplest model of chaotic inflation in supergravity. The non-gaussianity parameter f_NL in this scenario can take values in the observationally interesting range from O(10) to O(100). These values may be different in different parts of the universe. The regions where f_NL is particularly large form a curvaton web resembling a net of thick domain walls, strings, or global monopoles.

In a recent paper arXiv:0910.2230, Khoury and Steinhardt proposed a way to generate adiabatic cosmological perturbations with a nearly flat spectrum in a contracting Universe. To produce these perturbations they used a regime in which the equation of state exponentially rapidly changed during a short time interval. Leaving aside the singularity problem and the difficult question about the possibility to transmit these perturbations from a contracting Universe to the expanding phase, we will show that the methods used in arXiv:0910.2230 are inapplicable for the description of the cosmological evolution and of the process of generation of perturbations in this scenario.

We consider the possibility of generation of the seeds of primordial magnetic field on inflation and show that the effect of the back reaction of this field can be very important. Assuming that back reaction does not spoil inflation we find a rather strong restriction on the amplitude of the primordial seeds which could be generated on inflation. Namely, this amplitude recalculated to the present epoch cannot exceed $10^{-32}G$ in $Mpc$ scales. This field seems to be too small to be amplified to the observable values by galactic dynamo mechanism.

We discuss the gravitational waves (GW) in the context of vector inflation. We derive the action for tensor perturbations and find that tachyonic instabilities are present in most (but not all) of the inflationary models with large fields. In contrast, the stability of the small field inflation ($A_{\mu}A^{\mu}\ll\frac{1}{N}$) is ensured by the usual slow-roll conditions, where $N$ is the total number of fields. For example, the Coleman-Weinberg potential and the power-law inflation are always stable in the small fields limit with an approximately flat spectrum of GW. We also provide some examples which lead to a rapid decay of GW and predict the absence of tensor modes in the CMB.

We propose a scenario where inflation is driven by non-minimally coupled massive vector fields. In an isotropic homogeneous universe these fields behave in presicely the same way as a massive minimally coupled scalar field. Therefore our model is very similar to the model of chaotic inflation with scalar field. For vector fields the isotropy of expansion is achived either by considering a triplet of orthogonal vector fields or for the expense of $N$ randomly oriented vector fields. In the last case the substantial anisotropy of the expansion of order $1/\sqrt{N}$ survives until the end of inflation. The lightest vector fields might also force the late time acceleration of the Universe.

The new ekpyrotic scenario attempts to solve the singularity problem by involving violation of the null energy condition in a model which combines the ekpyrotic/cyclic scenario with the ghost condensate theory and the curvaton mechanism of production of adiabatic perturbations of metric. The Lagrangian of this theory, as well as of the ghost condensate model, contains a term with higher derivatives, which was added to the theory to stabilize its vacuum state. We found that this term may affect the dynamics of the cosmological evolution. Moreover, after a proper quantization, this term results in the existence of a new ghost field with negative energy, which leads to a catastrophic vacuum instability. We explain why one cannot treat this dangerous term as a correction valid only at small energies and momenta below some UV cut-off, and demonstrate the problems arising when one attempts to construct a UV completion of this theory.

The k-essence theories admit in general the superluminal propagation of the perturbations on classical backgrounds. We show that in spite of the superluminal propagation the causal paradoxes do not arise in these theories and in this respect they are not less safe than General Relativity.

In this short talk we review our results from the paper hep-th/0604075 with the main stress on the issues of causality and acoustic metric in eikonal approximation. In addition we correct the formula for the redshift of sound signals from our original work. We show that, if there exists a special kind of Born-Infeld type scalar field, then one can send information from inside a black hole. This information is encoded in perturbations of the field propagating in non-trivial scalar field backgrounds, which serve as a "new ether". Although the action of the theory is manifestly Lorentz-invariant the non-trivial solutions break this symmetry spontaneously, allowing, the superluminal propagation of perturbations with respect to this "new ether". We found the stationary solution for background, which describes the accretion of the scalar field onto a black hole. Examining the propagation of small perturbations around this solution we show that the signals emitted inside the Schwarzschild horizon can reach an observer located outside the black hole. Thus the accreting field forms a hydrodynamical analog of a black hole whose horizon is inside of the gravitational black hole drawing in the scalar field.

We show that if there exists a special kind of Born-Infeld type scalar field, then one can send information from inside a black hole. This information is encoded in perturbations of the field propagating in non-trivial scalar field backgrounds, which serves as a "new ether". Although the theory is Lorentz-invariant it allows, nevertheless, the superluminal propagation of perturbations with respect to the "new ether". We found the stationary solution for background, which describes the accretion of the scalar field onto a black hole. Examining the propagation of small perturbations around this solution we show the signals emitted inside the horizon can reach an observer located outside the black hole. We discuss possible physical consequences of this result.

Recently a non-inflationary mechanism of generation of scale-free cosmological perturbations of metric was proposed by Brandenberger, Nayeri, and Vafa in the context of the string gas cosmology. We discuss various problems of their model and argue that the cosmological perturbations of metric produced in this model have blue spectrum with a spectral index n = 5, which strongly disagrees with observations. We conclude that this model in its present form is not a viable alternative to inflationary cosmology.

We discuss nontrivial features of the large scale structure of the universe in the simplest curvaton model proposed in our paper astro-ph/9610219. The amplitude of metric perturbations in this model takes different values in different parts of the universe. The spatial distribution of the amplitude looks like a web consisting of exponentially large cells. Depending on the relation between the cell size l_0 and the scale of the horizon l_H, one may either live in a part of the universe dominated by gaussian perturbations (inside a cell with l_0 >> l_H), or in the universe dominated by nongaussian perturbations (for l_0 << l_H). We show that the curvaton contribution to the total amplitude of adiabatic density perturbations can be strongly suppressed if the energy density of the universe prior to the curvaton decay was dominated not by the classical curvaton field but by the curvaton particles produced during reheating. We describe the curvaton-inflaton transmutation effect: The same field in different parts of the universe may play either the role of the curvaton or the role of the inflaton. Finally, we discuss an interplay between the curvaton web and anthropic considerations in the string theory landscape.

We show that in theories with a nontrivial kinetic term the contribution of the gravitational waves to the CMB fluctuations can be substantially larger than that is naively expected in simple inflationary models. This increase of the tensor-to-scalar perturbation ratio leads to a larger B-component of the CMB polarization, thus making the prospects for future detection much more promising. The other important consequence of the considered model is a higher energy scale of inflation and hence higher reheating temperature compared to a simple inflation.

I review the motivation for the early-time cosmic acceleration stage in expanding universe and discuss simple inflationary scenarios. Preheating and reheating are considered in great detail. This is a sample chapter from my book "Physical Foundations of Cosmology" published by Cambridge University Press (2005).

We explain why it is so difficult and perhaps even impossible to increase the cosmological tensor-to-scalar perturbation ratio during the post-inflationary evolution of the universe. Nevertheless, contrary to some recent claims, tensor perturbations can be relatively large in the simplest inflationary models which do not violate any rules of modern quantum field theory.

I derive analytically the spectrum of the CMB fluctuations. The final result for C_l is presented in terms of elementary functions with an explicit dependence on the basic cosmological parameters. This result is in a rather good agreement with CMBFAST for a wide range of parameters around concordance model. This allows us to understand the physical reasons for dependence of the particular features of the CMB spectrum on the basic cosmological parameters and to estimate the possible accuracy of their determination. I also analyse the degeneracy of the spectrum with respect to certain combinations of the cosmological parameters.

I derive completely analytically the time evolution and final abundances of the light elements (up to Be-7) formed in the big-bang nucleosynthesis.This highlights an interesting physics taking place during the formation of light elements in the early universe.

I comment on the predictive power of inflation and recent CMB measurements.

Recently Hollands and Wald argued that inflation does not solve any of the major cosmological problems. We explain why we disagree with their arguments. They also proposed a new speculative mechanism of generation of density perturbations. We show that in their scenario the inhomogeneities responsible for the large scale structure observed today were generated at an epoch when the energy density of the hot universe was 10^{95} times greater than the Planck density. The only way to avoid this problem is to assume that there was a stage of inflation in the early universe.

Quintessence, a time-varying energy component that may account for the accelerated expansion of the universe, can be characterized by its equation of state and sound speed. In this paper, we show that if the quintessence density is at least one percent of the critical density at the surface of last scattering the cosmic microwave background anisotropy can distinguish between models whose sound speed is near the speed of light versus near zero, which could be useful in distinguishing competing candidates for dark energy.

We consider a cosmological model with non-Gaussian initial perturbations, which in principle could be generated in non-standard inflationary scenarios with two or more scalar fields. In particular we focus our attention on the model proposed by Linde & Mukhanov (1997) perturbations are quadratic in a Gaussian field. These perturbations, if they exist, have to be observable as a non-Gaussian distribution of the CMB signal on the sky. In order to efficiently pick up the non-Gaussian signal in CMB maps of degree resolution, one can use Minkowski Functionals and peak statistics. Our paper contains the theoretical predictions of the properties for Minkowski Functionals and distributions of peaks of the CMB anisotropy in the model with "squared" Gaussian statistics. Likelihood comparison of the four-year COBE DMR data to this non-Gaussian model and the standard Gaussian model does not select any of them as most likely. We also suggest an efficient algorithm for fast simulation of CMB maps on the whole sky. Using a cylindrical partition of the sphere, we rewrite the spherical harmonics analysis as a Fourier transform in flat space, which makes the problem accessible to numerically advantageous FFT methods.

We recently introduced the concept of "k-essence" as a dynamical solution for explaining naturally why the universe has entered an epoch of accelerated expansion at a late stage of its evolution. The solution avoids fine-tuning of parameters and anthropic arguments. Instead, k-essence is based on the idea of a dynamical attractor solution which causes it to act as a cosmological constant only at the onset of matter-domination. Consequently, k-essence overtakes the matter density and induces cosmic acceleration at about the present epoch. In this paper, we present the basic theory of k-essence and dynamical attractors based on evolving scalar fields with non-linear kinetic energy terms in the action. We present guidelines for constructing concrete examples and show that there are two classes of solutions, one in which cosmic acceleration continues forever and one in which the acceleration has finite duration.

We discuss the large-scale structure of the universe in inflationary cosmology and the implications that it may have for the long-term future of civilizations. Although each civilization is doomed to perish, it may be possible to transmit its accumulated knowledge to future civilizations. We consider several scenarios of this sort. If the cosmological constant is positive, it eventually dominates the universe and bubbles of inflationary phase begin to nucleate at a constant rate. Thermalized regions inside these inflating bubbles will give rise to new galaxies and civilizations. It is possible in principle to send a message to one of them. It might even be possible to send a device whose purpose is to recreate an approximation of the original civilization in the new region. However, the message or device will almost certainly be intercepted by black holes, which nucleate at a much higher rate than inflating bubbles. Formation of new inflating regions can also be triggered by gravitational collapse, but again the probability is low, and the number of attempts required for a positive outcome is enormous. The probability can be higher if the energy scale of inflation is closer to the Planck scale, but a high energy scale produces a tight bound on the amount of information that can be transmitted. One can try to avoid quantum tunneling altogether, but this requires a violation of quantum inequalities which constrain the magnitude of negative energy densities. However, the limits of validity of quantum inequalities are not clear, and future research may show that the required violation is in fact possible. Therein lies the hope for the future of civilizations.

Increasing evidence suggests that most of the energy density of the universe consists of a dark energy component with negative pressure, a ``cosmological constant" that causes the cosmic expansion to accelerate. In this paper, we address the puzzle of why this component comes to dominate the universe only recently rather than at some much earlier epoch. We present a class of theories based on an evolving scalar field where the explanation is based entirely on internal dynamical properties of the solutions. In the theories we consider, the dynamics causes the scalar field to lock automatically into a negative pressure state at the onset of matter-domination such that the present epoch is the earliest possible time, consistent with nucleosynthesis restrictions, when it can start to dominate.

We extend the theory of cosmological perturbations to the case when the ``matter'' Lagrangian is an arbitrary function of the scalar field and its first derivatives. In particular, this extension provides a unified description of known cases such as the usual scalar field and the hydrodynamical perfect fluid. In addition, it applies to the recently proposed k-inflation, which is driven by non-minimal kinetic terms in the Lagrangian. The spectrum of quantum fluctuations for slow-roll and power law k-inflation is calculated. We find, for instance, that the usual ``consistency relation'' between the tensor spectral index and the relative amplitude of scalar and tensor perturbations is modified. Thus, at least in principle, k-inflation is phenomenologically distinguishable from standard inflation.

It is shown that a large class of higher-order (i.e. non-quadratic) scalar kinetic terms can, without the help of potential terms, drive an inflationary evolution starting from rather generic initial conditions. In many models, this kinetically driven inflation (or "k-inflation" for short) rolls slowly from a high-curvature initial phase, down to a low-curvature phase and can exit inflation to end up being radiation-dominated, in a naturally graceful manner. We hope that this novel inflation mechanism might be useful in suggesting new ways of reconciling the string dilaton with inflation.

We draw attention to the possibility that inflation (i.e. accelerated expansion) might continue after the end of slow roll, during a period of fast oscillations of the inflaton field \phi . This phenomenon takes place when a mild non-convexity inequality is satisfied by the potential V(\phi). The presence of such a period of \phi-oscillation-driven inflation can substantially modify reheating scenarios. In some models the effect of these fast oscillations might be imprinted on the primordial perturbation spectrum at cosmological scales.

We derive a closed-form, analytical expression for the spectrum of long-wavelength density perturbations in inflationary models with two (or more) inflaton degrees of freedom that is valid in the slow-roll approximation. We illustrate several classes of potentials for which this expression reduces to a simple, algebraic expression.

We examine the theoretical foundations of standard methods for computing density perturbations in inflationary models. We find that: (1) the time-delay formalism (introduced by Guth and Pi, 1982) is only valid when inflation is well-described by the de Sitter solution and the equation-of-state is nearly unchanging; and, (2) the horizon-crossing/Bessel approximation extends to non-exponential inflation, but only if the equation-of-state is changing slowly. Integration of the gauge-invariant perturbation equations mode-by-mode is the only method reliable for general models. For models with rapidly varying equation-of-state, the correction leads to significantly different predictions for the microwave background anisotropy. An important corollary is that methods proposed for "reconstruction" of the inflaton potential from anisotropy data are unreliable for general models.

We study the effective energy-momentum tensor (EMT) for cosmological perturbations and formulate the gravitational back-reaction problem in a gauge invariant manner. We analyze the explicit expressions for the EMT in the cases of scalar metric fluctuations and of gravitational waves and derive the resulting equations of state. The formalism is applied to investigate the back-reaction effects in chaotic inflation. We find that for long wavelength scalar and tensor perturbations, the effective energy density is negative and thus counteracts any pre-existing cosmological constant. For scalar perturbations during an epoch of inflation, the equation of state is de Sitter-like.

In the simplest model of open inflation there are two inflaton fields decoupled from each other. One of them, the tunneling field, produces a first stage of inflation which prepares the ground for the nucleation of a highly symmetric bubble. The other, a free field, drives a second period of slow roll inflation inside the bubble. However, the second field also evolves during the first stage of inflation, which to some extent breaks the needed symmetry. We show that this generates large supercurvature anisotropies which, together with the results of Tanaka and Sasaki, rule out this class of simple models (unless, of course, $\Omega_0$ is sufficiently close to one.) The problem does not arise in modified models where the second field does not evolve in the first stage of inflation.

The back reaction of gravitational perturbations in a homogeneous background is determined by an effective energy-momentum tensor quadratic in the perturbations. We show that this nonlinear feedback effect is important in the case of long wavelength scalar perturbations in inflationary universe models. We also show how to solve an old problem concerning the gauge dependence of the effective energy-momentum tensor of perturbations.

We present a class of very simple inflationary models of two scalar fields which leads to nongaussian isothermal perturbations with "blue" spectrum, n > 1. One of the models is inspired by supersymmetric theories where light scalar fields naturally acquire masses of the order of the Hubble constant H during inflation. Another model presumes that one of the fields has a nonminimal interaction with gravity. By a slight modification of parameters of these models one can obtain either gaussian isothermal perturbations, or nongaussian adiabatic perturbations with n > 1.

We derive the effective energy-momentum tensor for cosmological perturbations and prove its gauge-invariance. The result is applied to study the influence of perturbations on the behaviour of the Friedmann background in inflationary Universe scenarios. We found that the back reaction of cosmological perturbations on the background can become important already at energies below the self-reproduction scale.

We compute the spectrum of scalar and tensor metric perturbations generated, as amplified vacuum fluctuations, during an epoch of dilaton-driven inflation of the type occurring naturally in string cosmology. In the tensor case the computation is straightforward while, in the scalar case, it is made delicate by the appearance of a growing mode in the familiar longitudinal gauge. In spite of this, a reliable perturbative calculation of perturbations far outside the horizon can be performed by resorting either to appropriate gauge invariant variables, or to a new coordinate system in which the growing mode can be "gauged down". The simple outcome of this complicated analysis is that both scalar and tensor perturbations exhibit nearly Planckian spectra, whose common "temperature" is related to some very basic parameters of the string-cosmology background.

A brief introduction to the gauge invariant classical and quantum theory of cosmological perturbations is given. The formalism is applied to inflationary Universe models and yields a consistent and unified description of the generation and evolution of fluctuations. A general formula for the amplitude of cosmological perturbations in inflationary cosmology is derived.

A theory of gravitation is constructed in which all homogeneous and isotropic solutions are nonsingular, and in which all curvature invariants are bounded. All solutions for which curvature invariants approach their limiting values approach de Sitter space. The action for this theory is obtained by a higher derivative modification of Einstein's theory. We expect that our model can easily be generalized to solve the singularity problem also for anisotropic cosmologies.

We derive a formula for the nonequilibrium entropy of a classical stochastic field in terms of correlation functions of this field. The formalism is then applied to define the entropy of gravitational perturbations (both gravitational waves and density fluctuations). We calculate this entropy in a specific cosmological model (the inflationary Universe) and find that on scales of interest in cosmology the entropy in both density perturbations and gravitational waves exceeds the entropy of statistical fluctuations of the microwave background. The nonequilibrium entropy discussed here is a measure of loss of information about the system. We discuss the origin of the entropy in our cosmological models and compare the definition of entropy in terms of correlation functions with the microcanonical definition in quantum statistical mechanics.

We propose a general definition of nonequilibrium entropy of a classical stochastic field. As an example of particular interest in cosmology we apply this definition to compute the entropy of density perturbations in an inflationary Universe. On the scales of structures in the Universe, the entropy of density perturbations dominates over the statistical fluctuations of the entropy of cosmic microwave photons, indicating the relevance of the entropy of density fluctuations for structure formation.