Polynomial $\hat{I}\pm$-attractors

Abstract

Inflationary $\hat{I}\pm$-attractor models can be naturally implemented in supergravity with hyperbolic geometry. They have stable predictions for observables, such as $n_s=1-2/N_e$, assuming that the potential in terms of the original geometric variables, as well as its derivatives, are not singular at the boundary of the hyperbolic disk, or half-plane. In these models, the potential in the canonically normalized inflaton field $\ddot{I}†$ has a plateau, which is approached exponentially fast at large $\ddot{I}†$. We call them exponential $\hat{I}\pm$-attractors. We present a closely related class of models, where the potential is not singular, but its derivative is singular at the boundary. The resulting inflaton potential is also a plateau potential, but it approaches the plateau polynomially. We call them polynomial $\hat{I}\pm$-attractors. Predictions of these two families of attractors completely cover the sweet spot of the Planck/BICEP/Keck data. The exponential ones are on the left, the polynomial are on the right.

12 pages, 3 figures, references added