Half-space solutions with $7/2$ frequency in the thin obstacle problem

For the thin obstacle problem in 3d, we show that half-space solutions form an isolated family in the space of $7/2$-homogeneous solutions. For a general solution with one blow-up profile in this family, we establish the rate of convergence to this profile. As a consequence, we obtain regularity of the free boundary near such contact points.

Quantifying entanglement in terms of an operational way

Quantifying entanglement is one of the most important tasks in the entanglement theory. In this paper, we establish entanglement monotones in terms of an operational approach, which is closely connected with the state conversion from pure states to the objective state by the local operations and classical communications (LOCC). It is shown that any good entanglement quantifier defined on pure states can induce an entanglement monotone for all density matrices. We especially show that our entanglement monotone is the maximal one among all that have the same form for pure states. In some particular cases, our proposed entanglement monotones turned to be equivalent to the convex roof construction, which hence gains an operational meaning. Some examples are given to demonstrate the different cases.

Fast generation of Cat states in Kerr nonlinear resonators via optimal adiabatic control

Macroscopic cat states have been widely studied to illustrate fundamental principles of quantum physics as well as their application in quantum information processing. In this paper, we propose a quantum speedup method for adiabatic creation of cat states in a Kerr nonlinear resonator via gradient-descent optimal adiabatic control. By simultaneously adiabatic tuning the the cavity detuning and driving field strength, the width of minimum energy gap between the target trajectory and non-adiabatic trajectory can be widen, which allows us to speed up the evolution along the adiabatic path. Compared with the previous proposal of preparing the cat state by only controlling two-photon pumping strength in a Kerr nonlinear resonator, our method can prepare the target state with much shorter time, as well as a high fidelity and a large non-classical volume. It is worth noting that the cat state prepared by our method is also robust against single-photon loss very well. Moreover, when our proposal has a large initial detuning, it will creates a large-size cat state successfully. This proposal of preparing cat states can be implemented in superconducting quantum circuits, which provides a quantum state resource for quantum information encoding and fault-tolerant quantum computing.

SimVLM: Simple Visual Language Model Pretraining with Weak Supervision

With recent progress in joint modeling of visual and textual representations, Vision-Language Pretraining (VLP) has achieved impressive performance on many multimodal downstream tasks. However, the requirement for expensive annotations including clean image captions and regional labels limits the scalability of existing approaches, and complicates the pretraining procedure with the introduction of multiple dataset-specific objectives. In this work, we relax these constraints and present a minimalist pretraining framework, named Simple Visual Language Model (SimVLM). Unlike prior work, SimVLM reduces the training complexity by exploiting large-scale weak supervision, and is trained end-to-end with a single prefix language modeling objective. Without utilizing extra data or task-specific customization, the resulting model significantly outperforms previous pretraining methods and achieves new state-of-the-art results on a wide range of discriminative and generative vision-language benchmarks, including VQA (+3.74% vqa-score), NLVR2 (+1.17% accuracy), SNLI-VE (+1.37% accuracy) and image captioning tasks (+10.1% average CIDEr score). Furthermore, we demonstrate that SimVLM acquires strong generalization and transfer ability, enabling zero-shot behavior including open-ended visual question answering and cross-modality transfer.

Voxel-wise Cross-Volume Representation Learning for 3D Neuron Reconstruction

Automatic 3D neuron reconstruction is critical for analysing the morphology and functionality of neurons in brain circuit activities. However, the performance of existing tracing algorithms is hinged by the low image quality. Recently, a series of deep learning based segmentation methods have been proposed to improve the quality of raw 3D optical image stacks by removing noises and restoring neuronal structures from low-contrast background. Due to the variety of neuron morphology and the lack of large neuron datasets, most of current neuron segmentation models rely on introducing complex and specially-designed submodules to a base architecture with the aim of encoding better feature representations. Though successful, extra burden would be put on computation during inference. Therefore, rather than modifying the base network, we shift our focus to the dataset itself. The encoder-decoder backbone used in most neuron segmentation models attends only intra-volume voxel points to learn structural features of neurons but neglect the shared intrinsic semantic features of voxels belonging to the same category among different volumes, which is also important for expressive representation learning. Hence, to better utilise the scarce dataset, we propose to explicitly exploit such intrinsic features of voxels through a novel voxel-level cross-volume representation learning paradigm on the basis of an encoder-decoder segmentation model. Our method introduces no extra cost during inference. Evaluated on 42 3D neuron images from BigNeuron project, our proposed method is demonstrated to improve the learning ability of the original segmentation model and further enhancing the reconstruction performance.

Rethinking Sampling Strategies for Unsupervised Person Re-identification

Unsupervised person re-identification (re-ID) remains a challenging task. While extensive research has focused on the framework design or loss function, we show in this paper that sampling strategy plays an equally important role. We analyze the reasons for differences in performance between various sampling strategies under the same framework and loss function. We suggest that deteriorated over-fitting is an important factor causing poor performance, and enhancing statistical stability can rectify this issue. Inspired by that, a simple yet effective approach is proposed, known as group sampling, which gathers groups of samples from the same class into a mini-batch. The model is thereby trained using normalized group samples, which helps to alleviate the effects associated with a single sample. Group sampling updates the pipeline of pseudo label generation by guaranteeing that samples are more efficiently divided into the correct classes. Group sampling regulates the representation learning process, which enhances statistical stability for feature representation in a progressive fashion. Qualitative and quantitative experiments on Market-1501, DukeMTMC-reID, and MSMT17 show that group sampling improves upon state-of-the-art methods by between 3.3%~6.1%. Code has been available at https://github.com/ucas-vg/GroupSampling.

Solid Source Metal-Organic Molecular Beam Epitaxy of Epitaxial RuO2

A seemingly simple oxide with a rutile structure, RuO2 has been shown to possess several intriguing properties ranging from strain-stabilized superconductivity to a strong catalytic activity. Much interest has arisen surrounding the controlled synthesis of RuO2 films but, unfortunately, utilizing atomically-controlled deposition techniques like molecular beam epitaxy (MBE) has been difficult due to the ultra-low vapor pressure and low oxidation potential of Ru. Here, we demonstrate the growth of epitaxial, single-crystalline RuO2 films on different substrate orientations using the novel solid-source metal-organic (MO) MBE. This approach circumvents these issues by supplying Ru using a pre-oxidized solid metal-organic precursor containing Ru. High-quality epitaxial RuO2 films with bulk-like room-temperature resistivity of 55 micro-ohm-cm were obtained at a substrate temperature as low as 300 C. By combining X-ray diffraction, transmission electron microscopy, and electrical measurements, we discuss the effect of substrate temperature, orientation, film thickness, and strain on the structure and electrical properties of these films. Our results illustrating the use of novel solid-source MOMBE approach paves the way to the atomic-layer controlled synthesis of complex oxides of stubborn metals, which are not only difficult to evaporate but also hard to oxidize.

Optimization of bus scheduling and bus-berth matching at curbside stops under connected vehicle environment

It is commonly seen that buses are blocked by the ones in front serving passengers and have to queue outside a curbside bus stop although there are vacant berths at the stop. The resultant bus delays degrade the service level of urban public transportation. A potential solution is to reschedule the arrivals of the buses at the stop for full utilization of the berths with the aid of connected vehicle technologies. This study proposes a mixed-integer linear programming model to optimize the scheduling of bus arrivals and the bus-berth matching at a curbside stop under connected vehicle environment. The objective is the minimization of the bus delays weighted by the number of passengers on the buses. Bus arrival times at the stop and the assignment of berths are optimized together with bus departure times from the stop. Bus punctuality is also taken into consideration. The proposed model could be applied dynamically to cater to time-varying traffic conditions. Numerical studies validate the advantages of the proposed model over the first-come-first-service strategy and the relaxed model without bus punctuality in terms of weighted bus delays and bus punctuality. Sensitivity analyses show that: 1) the proposed model is robust to the fluctuation of bus service time; and 2) a smaller number of berths may be preferred on condition that the bus demand does not exceed the stop capacity.

Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision

In this paper, we study the semi-supervised semantic segmentation problem via exploring both labeled data and extra unlabeled data. We propose a novel consistency regularization approach, called cross pseudo supervision (CPS). Our approach imposes the consistency on two segmentation networks perturbed with different initialization for the same input image. The pseudo one-hot label map, output from one perturbed segmentation network, is used to supervise the other segmentation network with the standard cross-entropy loss, and vice versa. The CPS consistency has two roles: encourage high similarity between the predictions of two perturbed networks for the same input image, and expand training data by using the unlabeled data with pseudo labels. Experiment results show that our approach achieves the state-of-the-art semi-supervised segmentation performance on Cityscapes and PASCAL VOC 2012. Code is available at https://git.io/CPS.

Learning Structures for Deep Neural Networks

In this paper, we focus on the unsupervised setting for structure learning of deep neural networks and propose to adopt the efficient coding principle, rooted in information theory and developed in computational neuroscience, to guide the procedure of structure learning without label information. This principle suggests that a good network structure should maximize the mutual information between inputs and outputs, or equivalently maximize the entropy of outputs under mild assumptions. We further establish connections between this principle and the theory of Bayesian optimal classification, and empirically verify that larger entropy of the outputs of a deep neural network indeed corresponds to a better classification accuracy. Then as an implementation of the principle, we show that sparse coding can effectively maximize the entropy of the output signals, and accordingly design an algorithm based on global group sparse coding to automatically learn the inter-layer connection and determine the depth of a neural network. Our experiments on a public image classification dataset demonstrate that using the structure learned from scratch by our proposed algorithm, one can achieve a classification accuracy comparable to the best expert-designed structure (i.e., convolutional neural networks (CNN)). In addition, our proposed algorithm successfully discovers the local connectivity (corresponding to local receptive fields in CNN) and invariance structure (corresponding to pulling in CNN), as well as achieves a good tradeoff between marginal performance gain and network depth.