We utilize machine learning to study the string landscape. Deep data dives and conjecture generation are proposed as useful frameworks for utilizing machine learning in the landscape, and examples of each are presented. A decision tree accurately predicts the number of weak Fano toric threefolds arising from reflexive polytopes, each of which determines a smooth F-theory compactification, and linear regression generates a previously proven conjecture for the gauge group rank in an ensemble of $\frac{4}{3} \times 2.96 \times 10^{755}$ F-theory compactifications. Logistic regression generates a new conjecture for when $E_{6}$ arises in the large ensemble of F-theory compactifications, which is then rigorously proven. This result may be relevant for the appearance of visible sectors in the ensemble. Through conjecture generation, machine learning is useful not only for numerics, but also for rigorous results.

We introduce network science as a framework for studying the string landscape. Two large networks of string geometries are constructed, where nodes are extra-dimensional six-manifolds and edges represent topological transitions between them. We show that a standard bubble cosmology model on the networks has late-time behavior determined by the largest eigenvector of $−(\mathbf{L}+\mathbf{D})$, where $\mathbf{L}$ and $\mathbf{D}$ are the Laplacian and degree matrices of the networks, which provides a dynamical mechanism for vacuum selection in the string landscape.

We provide the first estimate of the number of fine, regular, star triangulations of the four-dimensional reflexive polytopes, as classified by Kreuzer and Skarke (KS). This provides an upper bound on the number of Calabi-Yau threefold hypersurfaces in toric varieties. The estimate is performed with deep learning, specifically the novel equation learner (EQL) architecture. We demonstrate that EQL networks accurately predict numbers of triangulations far beyond the $h^{1,1}$ training region, allowing for reliable extrapolation. We estimate that number of triangulations in the KS dataset is $10^{10,505}$, dominated by the polytope with the highest $ h^{1,1} $ value.

We establish an orientifold Calabi-Yau threefold database for $h^{1,1}(X)≤6$ by considering non-trivial $\mathbb{Z}_2$ divisor exchange involutions, using a toric Calabi-Yau database. We first determine the topology for each individual divisor (Hodge diamond), then identify and classify the proper involutions which are globally consistent across all disjoint phases of the Kähler cone for each unique geometry. Each of the proper involutions will result in an orientifold Calabi-Yau manifold. Then we clarify all possible fixed loci under the pro per involution, thereby determining the locations of different types of $\mathcal{O}$-planes. It is shown that under the proper involutions, one typically ends up with a system of $\mathcal{O}3/\mathcal{O}7$-planes, and most of these will further admit naive Type IIB string vacua.The geometries with freely acting involutions are also determined. We further determine th e splitting of the Hodge numbers into odd/even parity in the orbifold limit. The final result is a class of orientifold Calabi-Yau threefolds with non-trivial odd class cohomology $h^{1,1 }_{−}(X/σ∗)≠0$.

The fields of astronomy and astrophysics are evolving and incorporating technologies to effectively view and explore 3D data visualizations, including Augmented Reality (AR). An analysis of the feasibility of using AR in journal publications for 3D visualizations took place two years ago, in 2023, when Adams et al. evaluated whether the perceived workload between AR and non-AR technologies was comparable. Given that the use of AR for astronomy journal publications was, and still is, in its infancy, the original study had to utilize data intended for K-12 education that had similar interactions and data types as a proxy for real-world data that could be visualized in future astronomy publications. In this paper, we present the results of a conceptual replication study of Adams et al.’s work to validate whether their findings hold with real astronomy stimuli. We found in our replication that many of the trends in the original study hold true, but that the workload experienced by participants was significantly higher under multiple conditions when using real-world data. Additionally, we found that the tradeoff between engagement and workload was as prevalent in the replication as it was in the original study. Our results provide a new framing for researchers to understand the tradeoffs of immersive visualization technologies and the increased workload of pairing these tools with complex, scientific stimuli. All Supplemental Material in our study is available at https://osf.io/j8urq/.

We describe the features of the Annular Eclipse Cosmic Data Story (DS), an online interactive resource that allows the public to visualize the October 14, 2023 annular solar eclipse from any location around the world, with a focus on North America. This resource is available online at: https://projects.cosmicds.cfa.harvard.edu/annular-eclipse-2023/. During the week of October 6-October 15, the Annular Eclipse DS received 2,000 views. Unless a user opted out of collection of evaluation data, we recorded the number of preset and user-selected locations they viewed and their responses to a multiple-choice map quiz. The group of users we received this data from are called the “evaluation cohort.” The user-selected locations were distributed across the entire US and beyond. On average, users in the evaluation cohort viewed the annular solar eclipse from 5 different locations. Within the evaluation cohort, 70% of users who attempted the map quiz arrived at the correct answer within two guesses.