Feature engineering has been the key to the success of many prediction models. However, the process is nontrivial and o en requires manual feature engineering or exhaustive searching. DNNs are able to automatically learn feature interactions; however, they generate all the interactions implicitly, and are not necessarily e cient in learning all types of cross features. In this paper, we propose the Deep & Cross Network (DCN) which keeps the bene ts of a DNN model, and beyond that, it introduces a novel cross network that is more e cient in learning certain bounded-degree feature interactions. In particular, DCN explicitly applies feature crossing at each layer, requires no manual feature engineering, and adds negligible extra complexity to the DNN model. Our experimental results have demonstrated its superiority over the state-of-art algorithms on the CTR prediction dataset and dense classi cation dataset, in terms of both model accuracy and memory usage.
A facile and efficient Pd/C-catalyzed carbonylation of both aliphatic and aromatic azides in the presence of amines is reported. Serving as the widely existed fragments in an array of biological pharmaceuticals, functionalized unsymmetrical ureas were straightforwardly synthesized by using readily available and cheap azides with amines under CO atmosphere, with the extrusion of N2 as the only byproduct. It was found that not only aryl azides but also benzyl and alkyl azides were suited for this methodology. Another feature of this procedure was the employment of a highly efficient palladium charcoal catalytic system.
Mitochondrial oxidative damage and dysfunction contribute to a wide range of human diseases. Considering the limitation of conventional antioxidants and that mitochondria are the main source of reactive oxygen species (ROS) which induce oxidative damage, mitochondria‐targeted antioxidants which can selectively block mitochondrial oxidative damage and prevent various types of cell death have been widely developed. As a lipophilic cation, triphenylphosphonium (TPP) has been commonly used in designing mitochondria‐targeted antioxidants. Conjugated with the TPP moiety, antioxidants can achieve more than 1000‐fold higher mitochondrial concentration depending on cell membrane potentials and mitochondrial membrane potentials. Herein we discuss the deficiencies of conventional antioxidants and the advantages of mitochondrial targeting, and review various types of TPP‐based mitochondria‐targeted antioxidants. These provide theoretical and background support for the design of new anti‐oxidant.
An efficient palladium-catalyzed cross-coupling reaction of azides with isocyanides is developed, providing a general synthetic route to unsymmetric carbodiimides with excellent yields. This method shows a broad substrate scope, including not only aryl azides, but also unactivated benzyl and alkyl azides. Furthermore, from readily available substrates, Pd-catalyzed coupling with a tandem amine insertion cascade to obtain unsymmetric trisubstituted guanidines has been achieved in a one-pot fashion.
Pd-catalyzed reactions of azides with CO to access an isocynate intermediate have been developed extensively in recent years. However, the catalytic carbonylation of sensitive acyl azides has not been reported. Herein, we report a simple Pdcatalyzed carbonylation reaction of acyl azides with broad substrate scope, high efficiency, and simple operation under mild conditions, which provides facile access to acyl ureas. In addition, a mechanistic study was carried out by both experiment and DFT calculation. Control experiments and kinetic study revealed that the real active palladium species were Pd(0). The result of kinetic study suggested that palladium catalyst, azide, and CO were all involved in the turnover-limiting step except for amine. Further DFT study suggested that an unprecedented five-membered palladacycle intermediate was the key intermediate in the carbonylation reaction.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.