Parabolic subgroups in FC-type Artin groups

Parabolic subgroups are the building blocks of Artin groups. This paper extends previous results of Cumplido, Gebhardt, Gonzales-Meneses and Wiest, known only for parabolic subgroups of finite type Artin groups, to parabolic subgroups of FC-type Artin groups. We show that the class of finite type parabolic subgroups is closed under intersection. We also study an analog of the curve complex for mapping class group constructed by Cumplido et al. using parabolic subgroups. We extend the construction of this complex, called the complex of parabolic subgroups, to FC-type Artin groups. We show that this simplicial complex is, in most cases, infinite diameter and conjecture that it is δ-hyperbolic.     

Reactivity of a N-Coordinated Germylene-Borane Complex: From Ge→B to Ge→Ga Coordination

Reactivity studies of the Ge^II→B complex L(Cl)Ge⋅BH₃ ( 1 ; L=2-Et₂NCH₂-4,6-tBu₂-C₆H₂) were performed to determine the effect on the Ge^II→B donation. N-coordinated compounds L(OtBu)Ge⋅BH₃ ( 2 ) and [LGe⋅BH₃]₂ ( 3 ) were prepared. The possible tuning of the Ge^II→B interaction was proved experimentally, yielding compounds 1-PPh₂-8-(LGe)-C₁₀H₆ ( 4 ) and L(Cl)Ge⋅GaCl₃ ( 5 ) without a Ge^II→B interaction. In 5 , an unprecedented Ge^II→Ga coordination was revealed. The experimental results were complemented by a theoretical study focusing on the bonding in 1 − 5 . The different strength of the Ge^II→E (E=B, Ga) donation was evaluated by using energy decomposition analysis. The basicity of different L(X)Ge groups through proton affinity is also assessed.     

Hyperparameter on-line learning of stochastic resonance based threshold networks

Aiming at training the feed-forward threshold neural network consisting of nondifferentiable activation functions, the approach of noise injection forms a stochastic resonance based threshold network that can be optimized by various gradient-based optimizers. The introduction of injected noise extends the noise level into the parameter space of the designed threshold network, but leads to a highly non-convex optimization landscape of the loss function. Thus, the hyperparameter on-line learning procedure with respective to network weights and noise levels becomes of challenge. It is shown that the Adam optimizer, as an adaptive variant of stochastic gradient descent, manifests its superior learning ability in training the stochastic resonance based threshold network effectively. Experimental results demonstrate the significant improvement of performance of the designed threshold network trained by the Adam optimizer for function approximation and image classification.     

On distance-regular Cayley graphs of generalized dicyclic groups

Let G be a generalized dicyclic group with identity 1. An inverse closed subset S of $G?\{1\}$ is called minimal if $〈S〉=G$ and there exists some $s\in S$ such that $〈S?\{s,s^{?1}\}〉\ne G$. In this paper, we characterize distance-regular Cayley graphs $\mathrm{Cay}(G,S)$ of G under the condition that S is minimal.     

Chemistry of Pentafluorosulfanyl Derivatives and Related Analogs: From Synthesis to Applications

Pentafluorosulfanyl (SF₅)-containing compounds and corresponding analogs are a highly valuable class of fluorine-containing building blocks owing to their unique properties. The reason for that is the set of peculiar and tremendously beneficial characteristics they can impart on molecules once introduced onto them. Despite this, their application in distinct scientific fields remains modest, given the extremely harsh reaction conditions needed to access such compounds. The recent synthetic approaches via S−F, and C−SF₅ bond formation as well as the use of SF₅-containing building blocks embody a “stairway-to-heaven” loophole in the synthesis of otherwise-inaccessible chemical scaffolds only a few years ago. Herein, we report and evaluate the properties of the SF₅ group and analogs, by summarizing synthetic methodologies available to access them as well as following applications in material science and medicinal chemistry since 2015.