In the last few years, deep learning has led to very good performance on a variety of problems, such as visual recognition, speech recognition and natural language processing. Among different types of deep neural networks, convolutional neural networks have been most extensively studied. Leveraging on the rapid growth in the amount of the annotated data and the great improvements in the strengths of graphics processor units, the research on convolutional neural networks has been emerged swiftly and achieved stateof-the-art results on various tasks. In this paper, we provide a broad survey of the recent advances in convolutional neural networks. We detailize the improvements of CNN on different aspects, including layer design, activation function, loss function, regularization, optimization and fast computation. Besides, we also introduce various applications of convolutional neural networks in computer vision, speech and natural language processing.

Document level sentiment classification remains a challenge: encoding the intrinsic relations between sentences in the semantic meaning of a document. To address this, we introduce a neural network model to learn vector-based document representation in a unified, bottom-up fashion. The model first learns sentence representation with convolutional neural network or long short-term memory. Afterwards, semantics of sentences and their relations are adaptively encoded in document representation with gated recurrent neural network. We conduct document level sentiment classification on four large-scale review datasets from IMDB and Yelp Dataset Challenge. Experimental results show that: (1) our neural model shows superior performances over several state-of-the-art algorithms; (2) gated recurrent neural network dramatically outperforms standard recurrent neural network in document modeling for sentiment classification. 1

In this work, we introduce Panoptic-DeepLab, a simple, strong, and fast system for panoptic segmentation, aiming to establish a solid baseline for bottom-up methods that can achieve comparable performance of two-stage methods while yielding fast inference speed. In particular, Panoptic-DeepLab adopts the dual-ASPP and dual-decoder structures specific to semantic, and instance segmentation, respectively. The semantic segmentation branch is the same as the typical design of any semantic segmentation model (e.g., DeepLab), while the instance segmentation branch is class-agnostic, involving a simple instance center regression. As a result, our single Panoptic-DeepLab simultaneously ranks first at all three Cityscapes benchmarks, setting the new state-of-art of 84.2% mIoU, 39.0% AP, and 65.5% PQ on test set. Additionally, equipped with MobileNetV3, Panoptic-DeepLab runs nearly in real-time with a single 1025 × 2049 image (15.8 frames per second), while achieving a competitive performance on Cityscapes (54.1 PQ% on test set). On Mapillary Vistas test set, our ensemble of six models attains 42.7% PQ, outperforming the challenge winner in 2018 by a healthy margin of 1.5%. Finally, our Panoptic-DeepLab also performs on par with several topdown approaches on the challenging COCO dataset. For the first time, we demonstrate a bottom-up approach could deliver state-of-the-art results on panoptic segmentation.

Hypoxia promotes not only the invasiveness of tumor cells, but also chemoresistance in cancer. Tumor associated macrophages (TAMs) residing at the site of hypoxic region of tumors have been known to cooperate with tumor cells, and promote proliferation and chemoresistance. Therefore, there is an urgent need for new strategies to alleviate tumor hypoxia and enhance chemotherapy response in solid tumors. Herein, we have taken advantage of high accumulation of TAMs in hypoxic regions of tumor and high reactivity of manganese dioxide nanoparticles (MnO2 NPs) toward hydrogen peroxide (H2O2) for the simultaneous production of O2 and regulation of pH to effectively alleviate tumor hypoxia by targeted delivery of MnO2 NPs to the hypoxic area. Furthermore, we also utilized the ability of hyaluronic acid (HA) modification in reprogramming anti-inflammatory, pro-tumoral M2 TAMs to pro-inflammatory, antitumor M1 macrophages to further enhance the ability of MnO2 NPs to lessen tumor hypoxia and modulate chemoresistance. The HA-coated, mannanconjugated MnO2 particle (Man-HA-MnO2) treatment significantly increased tumor oxygenation and down-regulated hypoxia-inducible factor-1 α (HIF-1α) and vascular endothelial growth factor (VEGF) in the tumor. Combination treatment of the tumors with Man-HA-MnO2 NPs and doxorubicin significantly increased apparent diffusion coefficient (ADC) values of breast tumor, inhibited tumor growth and tumor cell proliferation as compared with chemotherapy alone. In addition, the reaction of Man-HA-MnO2 NPs toward endogenous H2O2 highly enhanced T1- and T2-MRI performance for tumor imaging and detection.

We propose a weakly supervised temporal action localization algorithm on untrimmed videos using convolutional neural networks. Our algorithm learns from video-level class labels and predicts temporal intervals of human actions with no requirement of temporal localization annotations. We design our network to identify a sparse subset of key segments associated with target actions in a video using an attention module and fuse the key segments through adaptive temporal pooling. Our loss function is comprised of two terms that minimize the video-level action classification error and enforce the sparsity of the segment selection. At inference time, we extract and score temporal proposals using temporal class activations and class-agnostic attentions to estimate the time intervals that correspond to target actions. The proposed algorithm attains state-of-the-art results on the THUMOS14 dataset and outstanding performance on ActivityNet1.3 even with its weak supervision.