A Search for Multiplanet Systems with TESS Using a Bayesian N-body Retrieval and Machine Learning

Transiting exoplanets in multiplanet systems exhibit non-Keplerian orbits as a result of the gravitational influence from companions, which can cause the times and durations of transits to vary. The amplitude and periodicity of the transit time variations are characteristic of the perturbing planet’s mass and orbit. The objects of interest from the Transiting Exoplanet Survey Satellite (TESS) are analyzed in a uniform way to search for transit timing variations (TTVs) with sectors 1-3 of data. Due to the volume of targets in the TESS candidate list, artificial intelligence is used to expedite the search for planets by vetting nontransit signals prior to characterizing the light-curve time series. The residuals of fitting a linear orbit ephemeris are used to search for TTVs. The significance of a perturbing planet is assessed by comparing the Bayesian evidence between a linear and nonlinear ephemeris, which is based on an N-body simulation. Nested sampling is used to derive posterior distributions for the N-body ephemeris and in order to expedite convergence, custom priors are designed using machine learning. A dual-input, multi-output convolutional neural network is designed to predict the parameters of a perturbing body given the known parameters and measured perturbation (O - C). There is evidence for three new multiplanet candidates (WASP-18, WASP-126, TOI 193) with nontransiting companions using the two-minute cadence observations from TESS. This approach can be used to identify stars in need of longer radial velocity and photometric follow-up than those already performed.