Recent breakthroughs in Neural Architectural Search (NAS) have achieved state-of-the-art performance in many tasks such as image classification and language understanding. However, most existing works only optimize for model accuracy and largely ignore other important factors imposed by the underlying hardware and devices, such as latency and energy, when making inference. In this paper, we first introduce the problem of NAS and provide a survey on recent works. Then we deep dive into two recent advancements on extending NAS into multiple-objective frameworks: MONAS [1] and DPP-Net [2] . Both MONAS and DPP-Net are capable of optimizing accuracy and other objectives imposed by devices, searching for neural architectures that can be best deployed on a wide spectrum of devices: from embedded systems and mobile devices to workstations. Experimental results are poised to show that architectures found by MONAS and DPP-Net achieves Pareto optimality w.r.t the given objectives for various devices.
Neural Architecture SearchIn this section, we survey the recent literatures on NAS and summarize them into four categories: (a) reinforcementlearning based methods, (b) evolutionary-algorithm based methods, (c) search acceleration, and (d) multi-objective search. Table. 1 provides the overview and comparisons among these literatures.
Problem DefinitionGenerally, the problem of neural architecture search can be formulated into two sub-problems: design "Search Space" and "Search Algorithm" [28] .Search Space As its name suggests, search space represents a set of possible neural networks available to be searched over. Usually, a search space has a numerical representation that contains:• Structure of a neural network, such as the depth of a neural net (i.e., the number of hidden layers) and the width of a particular hidden layer.• Configurations, such as operation/connection types, kernel size, the number of filters.
Human mesenchymal stem cells (MSCs) modified by targeting DNA hypermethylation of genes in the Salvador/Warts/Hippo pathway were induced to differentiate into neuronal cells in vitro. The differentiated cells secreted a significant level of brain-derived neurotrophy factor (BDNF) and the expression of BDNF receptor tyrosine receptor kinase B (TrkB) correlated well with the secretion of BDNF. In the differentiating cells, CREB was active after the binding of growth factors to induce phosphorylation of ERK in the MAPK/ERK pathway. Downstream of phosphorylated CREB led to the functional maturation of differentiated cells and secretion of BDNF, which contributed to the sustained expression of pERK and pCREB. In summary, both PI3K/Akt and MAPK/ERK signaling pathways play important roles in the neuronal differentiation of MSCs. The main function of the PI3K/Akt pathway is to maintain cell survival during neural differentiation; whereas the role of the MAPK/ERK pathway is probably to promote the maturation of differentiated MSCs. Further, cellular levels of protein kinase C epsilon type (PKC-ε) and kinesin heavy chain (KIF5B) increased with time of induction, whereas the level of NME/NM23 nucleoside diphosphate kinase 1 (Nm23-H1) decreased during the time course of differentiation. The correlation between PKC-ε and TrkB suggested that there is cross-talk between PKC-ε and the PI3K/Akt signaling pathway.
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.