51�������

Abstract:

Recent JWST/NIRCam imaging taken for the ultra-deep UNCOVER program reveals a very red dropout object at z phot ≃ 7.6, triply imaged by the galaxy cluster A2744 (z d = 0.308). All three images are very compact, i.e., unresolved, with a delensed size upper limit of r e ≲ 35 pc. The images have apparent magnitudes of m F444W ∼ 25−26 AB, and the magnification-corrected absolute UV magnitude of the source is M UV,1450 = −16.81 ± 0.09. From the sum of observed fluxes and from a spectral energy distribution (SED) analysis, we obtain estimates of the bolometric luminosities of the source of L bol ≳ 1043 erg s−1 and L bol ∼ 1044–1046 erg s−1, respectively. Based on its compact, point-like appearance, its position in color–color space, and the SED analysis, we tentatively conclude that this object is a UV-faint dust-obscured quasar-like object, i.e., an active galactic nucleus at high redshift. We also discuss other alternative origins for the object’s emission features, including a massive star cluster, Population III, supermassive, or dark stars, or a direct-collapse black hole. Although populations of red galaxies at similar photometric redshifts have been detected with JWST, this object is unique in that its high-redshift nature is corroborated geometrically by lensing, that it is unresolved despite being magnified—and thus intrinsically even more compact—and that it occupies notably distinct regions in both size–luminosity and color–color space. Planned UNCOVER JWST/NIRSpec observations, scheduled in Cycle 1, will enable a more detailed analysis of this object

Abstract:

When choosing between competing symbolic models for a data set, a human will naturally prefer the “simpler” expression or the one which more closely resembles equations previously seen in a similar context. This suggests a non-uniform prior on functions, which is, however, rarely considered within a symbolic regression (SR) framework. In this paper we develop methods to incorporate detailed prior information on both functions and their parameters into SR. Our prior on the structure of a function is based on a ngram language model, which is sensitive to the arrangement of operators relative to one another in addition to the frequency of occurrence of each operator. We also develop a formalism based on the Fractional Bayes Factor to treat numerical parameter priors in such a way that models may be fairly compared though the Bayesian evidence, and explicitly compare Bayesian, Minimum Description Length and heuristic methods for model selection. We demonstrate the performance of our priors relative to literature standards on benchmarks and a real-world dataset from the field of cosmology.