5分で分かる!? 有名論文ナナメ読み：Jacob Devlin et al. : BERT : Pre-training of Deep Bidirectional Transformers for Language Understanding

Deep learning (DL) techniques have obtained remarkable achievements on various tasks, such as image recognition, object detection, and language modeling. However, building a high-quality DL system for a specific task highly relies on human expertise, hindering its wide application. Meanwhile, automated machine learning (AutoML) is a promising solution for building a DL system without human assistance and is being extensively studied. This paper presents a comprehensive and up-to-date review of the state-of-the-art (SOTA) in AutoML. According to the DL pipeline, we introduce AutoML methods – covering data preparation, feature engineering, hyperparameter optimization, and neural architecture search (NAS) – with a particular focus on NAS, as it is currently a hot sub-topic of AutoML. We summarize the representative NAS algorithms’ performance on the CIFAR-10 and ImageNet datasets and further discuss the following subjects of NAS methods: one/two-stage NAS, one-shot NAS, joint hyperparameter and architecture optimization, and resource-aware NAS. Finally, we discuss some open problems related to the existing AutoML methods for future research.

AutoML: A survey of the state-of-the-art

Summary Bacterial cellulose, an exopolysaccharide produced by some bacteria, has unique structural and mechanical properties and is highly pure as compared to plant cellulose. This article presents a critical review of the available information on the bacterial cellulose with special emphasis on its fermentative production and applications. Information on the biosynthetic pathway of bacterial cellulose, enzymes and precursors involved in bacterial cellulose synthesis has been specified. Characteristics of bacterial cellulose with respect to its structure and physicochemical properties are discussed. Current and potential applications of bacterial cellulose in food, pharmaceutical and other industries are also presented.

/pdf/microbial-cellulose-fermentative-production-and-applications-bqh4jn98f9.pdf

Microbial Cellulose: Fermentative Production and Applications

Facial expression recognition in the wild is challenging due to various unconstrained conditions. Although existing facial expression classifiers have been almost perfect on analyzing constrained frontal faces, they fail to perform well on partially occluded faces that are common in the wild. In this paper, we propose a convolution neutral network (CNN) with attention mechanism (ACNN) that can perceive the occlusion regions of the face and focus on the most discriminative un-occluded regions. ACNN is an end-to-end learning framework. It combines the multiple representations from facial regions of interest (ROIs). Each representation is weighed via a proposed gate unit that computes an adaptive weight from the region itself according to the unobstructedness and importance. Considering different RoIs, we introduce two versions of ACNN: patch-based ACNN (pACNN) and global–local-based ACNN (gACNN). pACNN only pays attention to local facial patches. gACNN integrates local representations at patch-level with global representation at image-level. The proposed ACNNs are evaluated on both real and synthetic occlusions, including a self-collected facial expression dataset with real-world occlusions, the two largest in-the-wild facial expression datasets (RAF-DB and AffectNet) and their modifications with synthesized facial occlusions. Experimental results show that ACNNs improve the recognition accuracy on both the non-occluded faces and occluded faces. Visualization results demonstrate that, compared with the CNN without Gate Unit, ACNNs are capable of shifting the attention from the occluded patches to other related but unobstructed ones. ACNNs also outperform other state-of-the-art methods on several widely used in-the-lab facial expression datasets under the cross-dataset evaluation protocol.

Occlusion Aware Facial Expression Recognition Using CNN With Attention Mechanism

Utilization of bacterial cellulose in food

Bacterial cellulose (BC) production by Acetobacter xylinum NUST4.1 was carried out in the shake flask and in a stirred-tank reactor by means of adding sodium alginate (NaAlg) into the medium. When 0.04% (w/v) NaAlg was added in the shake flask, BC production reached 6.0 g/l and the terminal yield of the cellulose was 27% of the total sugar initially added, compared with 3.7 g/l and 24% in the control, respectively. The variation between replicates in all determinations was less than 5%. During the cultivation in the stirred-tank reactor, the addition of NaAlg changed the morphology of cellulose from the irregular clumps and fibrous masses entangled in the internals to discrete masses dispersing into the broth, which indicates that NaAlg hinders formation of large clumps of BC, and enhances cellulose yield. Because the structure of cellulose is changed depending on the culture condition such as additives, structural characteristics of BC produced in the NaAlg-free and NaAlg medium are compared using scanning electron microscopy (SEM), fourier transform infrared spectroscopy (FT-IR), X-ray diffraction (XRD). SEM photographs show some differences in reticulated structures and ribbon width and FT-IR spectra indicate that there is the hydrogen bonding interaction between BC and NaAlg, then X-ray diffraction (XRD) analysis reveals that BC produced with NaAlg-added has a lower crystallinity and a smaller crystalline size. The results show that enhanced yields and modification of cellulose structure occur in the presence of NaAlg.

Effect of addition of sodium alginate on bacterial cellulose production by Acetobacter xylinum

Facial expression recognition in the wild is challenging due to various un-constrained conditions. Although existing facial expression classifiers have been almost perfect on analyzing constrained frontal faces, they fail to perform well on partially occluded faces that are common in the wild. In this paper, we propose an end-to-end trainable Patch-Gated Convolution Neutral Network (PG-CNN) that can automatically percept the occluded region of the face and focus on the most discriminative un-occluded regions. To determine the possible regions of interest on the face, PG-CNN decomposes an intermediate feature map into several patches according to the positions of related facial landmarks. Then, via a proposed Patch-Gated Unit, PG-CNN reweighs each patch by the unobstructed-ness or importance that is computed from the patch itself. The proposed PG-CNN is evaluated on two largest in-the-wild facial expression datasets (RAF-DB and AffectNet) and their modifications with synthesized facial occlusions. Experimental results show that PG-CNN improves the recognition accuracy on both the original faces and faces with synthesized occlusions. Visualization results demonstrate that, compared with the CNN without Patch-Gated Unit, PG-CNN is capable of shifting the attention from the occluded patch to other related but unobstructed ones. Experiments also show that PG-CNN outperforms other state-of-the-art methods on several widely used in-the-lab facial expression datasets under the cross-dataset evaluation protocol.

Patch-Gated CNN for Occlusion-aware Facial Expression Recognition

In this paper, we aim to learn discriminative representation for facial action unit (AU) detection from large amount of videos without manual annotations. Inspired by the fact that facial actions are the movements of facial muscles, we depict the movements as the transformation between two face images in different frames and use it as the self-supervisory signal to learn the representations. However, under the uncontrolled condition, the transformation is caused by both facial actions and head motions. To remove the influence by head motions, we propose a Twin-Cycle Autoencoder (TCAE) that can disentangle the facial action related movements and the head motion related ones. Specifically, TCAE is trained to respectively change the facial actions and head poses of the source face to those of the target face. Our experiments validate TCAE's capability of decoupling the movements. Experimental results also demonstrate that the learned representation is discriminative for AU detection, where TCAE outperforms or is comparable with the state-of-the-art self-supervised learning methods and supervised AU detection methods.

/pdf/self-supervised-representation-learning-from-videos-for-4v7cu1f1uw.pdf

Self-Supervised Representation Learning From Videos for Facial Action Unit Detection

Frame reconstruction (current or future frame) based on Auto-Encoder (AE) is a popular method for video anomaly detection. With models trained on the normal data, the reconstruction errors of anomalous scenes are usually much larger than those of normal ones. Previous methods introduced the memory bank into AE, for encoding diverse normal patterns across the training videos. However, they are memory-consuming and cannot cope with unseen new scenarios in the testing data. In this work, we propose a dynamic prototype unit (DPU) to encode the normal dynamics as prototypes in real time, free from extra memory cost. In addition, we introduce meta-learning to our DPU to form a novel few-shot normalcy learner, namely Meta-Prototype Unit (MPU). It enables the fast adaption capability on new scenes by only consuming a few iterations of update. Extensive experiments are conducted on various benchmarks. The superior performance over the state-of-the-art demonstrates the effectiveness of our method. Our code is available at https://github.com/ktr-hubrt/MPN/.

/pdf/learning-normal-dynamics-in-videos-with-meta-prototype-1h0t954s1l.pdf

Yong Li

Papers

Occlusion Aware Facial Expression Recognition Using CNN With Attention Mechanism

Effect of addition of sodium alginate on bacterial cellulose production by Acetobacter xylinum

Patch-Gated CNN for Occlusion-aware Facial Expression Recognition

Self-Supervised Representation Learning From Videos for Facial Action Unit Detection

Learning Normal Dynamics in Videos with Meta Prototype Network