Search for: All records

Due to the limited availability of actual large-scale datasets, realistic synthetic trajectory data play a crucial role in various research domains, including spatiotemporal data mining and data management, and domain-driven research related to transportation planning and urban analytics. Existing generation methods rely on predefined heuristics and cannot learn the unknown underlying generative mechanisms. This work introduces two end-to-end approaches for trajectory generation. The first approach comprises deep generative VAE-like models that factorize global and local semantics (habits vs. random routing change). We further enhance this approach by developing novel inference strategies based on variational inference and constrained optimization to ensure the validity of spatiotemporal aspects. This novel deep neural network architecture implements generative and inference models with dynamic latent priors. The second approach introduces a language model (LM) inspired generation as another benchmarking and foundational approach. The LM-inspired approach conceptualizes trajectories as sentences with the aim of predicting the likelihood of subsequent locations on a trajectory, given the locations as context. As a result, the LM-inspired approach implicitly learns the inherent spatiotemporal structure and other embedded semantics within the trajectories. These proposed methods demonstrate substantial quantitative and qualitative improvements over existing approaches, as evidenced by extensive experimental evaluations. 

Enantioselective protonation is a versatile approach to the construction of tertiary α-stereocenters, which are common structural motifs in various natural products and biologically relevant compounds. Herein we report a mild access to these chiral centers using cooperative gold(I) catalysis. From cyclic ketone enol carbonates, this asymmetric catalysis provides highly enantioselective access to cyclic ketones featuring an α tertiary chiral center, including challenging 2-methylsuberone. In combination with the gold-catalyzed formation of cyclopentadienyl carbonates in a one-pot, two-step process, this chemistry enables expedient access to synthetically versatile α′-chiral cyclopentenones with excellent enantiomeric excesses from easily accessible enynyl carbonate substrates. 

Abstract Cyclopentene rings possessing a chiral quaternary center are important structural motifs found in various natural products. In this work, we disclose expedient and efficient access to this class of synthetically valuable structuresviahighly enantioselective desymmetrization of prochiral propargylic alcohols. The efficient chirality induction in this asymmetric gold catalysis is achievedviatwo‐point bindings between a gold catalyst featuring a bifunctional phosphine ligand and the substrate homopropargylic alcohol moiety—an H‐bonding interaction between the substrate HO group and a ligand phosphine oxide moiety and the gold‐alkyne complexation. The propargylic alcohol substrates can be prepared readilyviapropargylation of enoate and ketone precursors. In addition to monocyclic cyclopentenes, spirocyclic and bicyclic ones are formed with additional neighboring chiral centers of flexible stereochemistry in addition to the quaternary center. This work represents rare gold‐catalyzed highly enantioselective cycloisomerization of 1,5‐enynes. Density functional theory (DFT) calculations support the chirality induction model and suggest that the rate acceleration enabled by the bifunctional ligand can be attributed to a facilitated protodeauration step at the end of the catalysis.