Face recognition has made extraordinary progress owing to the advancement of deep convolutional neural networks (CNNs). The central task of face recognition, including face verification and identification, involves face feature discrimination. However, the traditional softmax loss of deep CNNs usually lacks the power of discrimination. To address this problem, recently several loss functions such as center loss, large margin softmax loss, and angular softmax loss have been proposed. All these improved losses share the same idea: maximizing inter-class variance and minimizing intra-class variance. In this paper, we propose a novel loss function, namely large margin cosine loss (LMCL), to realize this idea from a different perspective. More specifically, we reformulate the softmax loss as a cosine loss by L 2 normalizing both features and weight vectors to remove radial variations, based on which a cosine margin term is introduced to further maximize the decision margin in the angular space. As a result, minimum intra-class variance and maximum inter-class variance are achieved by virtue of normalization and cosine decision margin maximization. We refer to our model trained with LMCL as CosFace.
Hydrogels have exhibited remarkable benefits in drug delivery such as local delivery, days or even weeks of continuous drug release with improved bioavailability, and minimized adverse effects.Here we report a polydopamine (PDA) nanoparticle-knotted poly(ethylene glycol) (PEG) hydrogel for on-demand drug delivery and combined chemo-photothermal therapy. Anticancer drugs such as 7-ethyl-10-hydroxycamptothecin (SN38) loaded on PDA nanoparticles via π−π interaction in the gel exhibit minimal leakage at physiological conditions and could be released in an on-demand fashion upon near-infrared light exposure. The hydrogel shows excellent biocompatibility and does not induce any foreign-body reaction over a four-month implantation. The in vivo results demonstrate that the PDA nanoparticle-knotted PEG hydrogel loaded with SN38 could efficiently suppress tumor growth by a combined chemo-photothermal therapy. This smart hydrogel would benefit a series of local treatments for diverse diseases.
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.