Search

Xiaoyu Wu,1,2 Chun Liu,2 Yuqing Jiang,2 Ting Dai,1,2 Linxiang Zhang,1,2 Jiafeng Wang,1,2 Hongbin Zhao1,2 1Gansu Provincial Maternity and Child-Care Hospital, Gansu Provincial Central Hospital, Gansu, People’s Republic of China; 2Changzhou Medical Center, The Affiliated Changzhou Second People’s Hospital of Nanjing Medical University, Changzhou, People’s Republic of ChinaCorrespondence: Hongbin Zhao, Gansu Provincial Maternity and Child-care Hospital, Gansu Provincial Central Hospital, Gansu, 730070, People’s Republic of China, Email zhao761032@163.comBackground: Salidroside (SAL) is the most effective component of Rhodiola rosea, a traditional Chinese medicine. Cryptotanshinone (CT) is the main fat-soluble extract of Salvia miltiorrhiza, exhibiting considerable potential for application in osteogenesis. Herein, a polycaprolactone/gelatin nanofiber membrane loaded with CT and SAL (PSGC membrane) was successfully fabricated via coaxial electrospinning and characterized.Methods and Results: This membrane capable of sustained and controlled drug release was employed in this study. Co-culturing the membrane with bone marrow mesenchymal stem cells and human umbilical vein endothelial cells revealed excellent biocompatibility and demonstrated osteogenic and angiogenic capabilities. Furthermore, drug release from the PSGC membrane activated the Wnt/β-catenin signaling pathway and promoted osteogenic differentiation and vascularization. Evaluation of the membrane’s vascularization and osteogenic capacities involved transplantation onto a rat’s subcutaneous area and assessing rat cranium defects for bone regeneration, respectively. Microcomputed tomography, histological tests, immunohistochemistry, and immunofluorescence staining confirmed the membrane’s outstanding angiogenic capacity two weeks post-operation, with a higher incidence of osteogenesis observed in rat cranial defects eight weeks post-surgery.Conclusion: Overall, the SAL- and CT-loaded coaxial ele

Diffusion Models (DMs) have evolved into advanced image generation tools, especially for few-shot generation where a pretrained model is fine-tuned on a small set of images to capture a specific style or object. Despite their success, concerns exist about potential copyright violations stemming from the use of unauthorized data in this process. In response, we present Contrasting Gradient Inversion for Diffusion Models (CGI-DM), a novel method featuring vivid visual representations for digital copyright authentication. Our approach involves removing partial information of an image and recovering missing details by exploiting conceptual differences between the pretrained and fine-tuned models. We formulate the differences as KL divergence between latent variables of the two models when given the same input image, which can be maximized through Monte Carlo sampling and Projected Gradient Descent (PGD). The similarity between original and recovered images serves as a strong indicator of potential infringements. Extensive experiments on the WikiArt and Dreambooth datasets demonstrate the high accuracy of CGI-DM in digital copyright authentication, surpassing alternative validation techniques. Code implementation is available at https://github.com/Nicholas0228/Revelio., Comment: Accepted by CVPR 2024

Few-shot fine-tuning of Diffusion Models (DMs) is a key advancement, significantly reducing training costs and enabling personalized AI applications. However, we explore the training dynamics of DMs and observe an unanticipated phenomenon: during the training process, image fidelity initially improves, then unexpectedly deteriorates with the emergence of noisy patterns, only to recover later with severe overfitting. We term the stage with generated noisy patterns as corruption stage. To understand this corruption stage, we begin by theoretically modeling the one-shot fine-tuning scenario, and then extend this modeling to more general cases. Through this modeling, we identify the primary cause of this corruption stage: a narrowed learning distribution inherent in the nature of few-shot fine-tuning. To tackle this, we apply Bayesian Neural Networks (BNNs) on DMs with variational inference to implicitly broaden the learned distribution, and present that the learning target of the BNNs can be naturally regarded as an expectation of the diffusion loss and a further regularization with the pretrained DMs. This approach is highly compatible with current few-shot fine-tuning methods in DMs and does not introduce any extra inference costs. Experimental results demonstrate that our method significantly mitigates corruption, and improves the fidelity, quality and diversity of the generated images in both object-driven and subject-driven generation tasks., Comment: Preprint. Under review

Rabies is a zoonotic neurological disease caused by the rabies virus (RABV) that is fatal to humans and animals. While several post-infection treatment have been suggested, developing more efficient and innovative antiviral methods are necessary due to the limitations of current therapeutic approaches. To address this challenge, a strategy combining photodynamic therapy and immunotherapy, using a photosensitizer (TPA-Py-PhMe) with high type I and type II reactive oxygen species (ROS) generation ability is proposed. This approach can inactivate the RABV by killing the virus directly and activating the immune response. At the cellular level, TPA-Py-PhMe can reduce the virus titer under preinfection prophylaxis and postinfection treatment, with its antiviral effect mainly dependent on ROS and pro-inflammatory factors. Intriguingly, when mice are injected with TPA-Py-PhMe and exposed to white light irradiation at three days post-infection, the onset of disease is delayed, and survival rates improved to some extent. Overall, this study shows that photodynamic therapy and immunotherapy open new avenues for future antiviral research.

This paper proposes a hierarchical robot manipulation control scheme for the manipulator pushing task. Considering the contact surface and the object mass parameters are generally unknown to the pushing task, a hierarchical control strategy is designed that enables the robot manipulator to perform the manipulation pushing task without any prior knowledge of the objects mass and friction. Specifically, we first present an adaptive controller to achieve the desired interaction force exerted on the end of the manipulator even without knowing the object mass and friction contact surface. We then introduce a unified robust adaptive controller that takes into account the desired interaction manipulation force in our robot manipulator dynamics. A robust inverse dynamic controller combining a barrier Lyapunov function is included to prevent constraint violation, which can effectively keep the object motion along the desired trajectory while regulating the interaction force between the robot manipulator and the object. Furthermore, an adaptive unit control gain is designed to make the manipulator adapt to follow the perturbation variations without being overestimated. An experiment has been performed to test the effectiveness and to demonstrate the guaranteed performance of the proposed methods.

Recently, Diffusion Models (DMs) boost a wave in AI for Art yet raise new copyright concerns, where infringers benefit from using unauthorized paintings to train DMs to generate novel paintings in a similar style. To address these emerging copyright violations, in this paper, we are the first to explore and propose to utilize adversarial examples for DMs to protect human-created artworks. Specifically, we first build a theoretical framework to define and evaluate the adversarial examples for DMs. Then, based on this framework, we design a novel algorithm, named AdvDM, which exploits a Monte-Carlo estimation of adversarial examples for DMs by optimizing upon different latent variables sampled from the reverse process of DMs. Extensive experiments show that the generated adversarial examples can effectively hinder DMs from extracting their features. Therefore, our method can be a powerful tool for human artists to protect their copyright against infringers equipped with DM-based AI-for-Art applications. The code of our method is available on GitHub: https://github.com/mist-project/mist.git., Comment: Accepted by ICML2023 (Oral)

Xiaoyu Wu,1,* Danfei Ji,1,* Zefan Wang,1,* Wenhua Yu,2 Quan Du,2 Wei Hu,3 Yongke Zheng,3 Xiaoqiao Dong,2 Fanghui Chen4 1The Fourth School of Clinical Medicine, Zhejiang Chinese Medical University, Hangzhou, People’s Republic of China; 2Department of Neurosurgery, Affiliated Hangzhou First People’s Hospital, Zhejiang University School of Medicine, Hangzhou, People’s Republic of China; 3Department of Intensive Care Unit, Affiliated Hangzhou First People’s Hospital, Zhejiang University School of Medicine, Hangzhou, People’s Republic of China; 4Emergency Department, Affiliated Hangzhou First People’s Hospital, Zhejiang University School of Medicine, Hangzhou, People’s Republic of China*These authors contributed equally to this workCorrespondence: Xiaoqiao Dong; Fanghui Chen, Email dxqhyy@163.com; wangxiaoleihzsy@163.comBackground: NADPH oxidase 2 (NOX2) is highly expressed in injured brain tissues. We determined serum NOX2 levels of aneurysmal subarachnoid hemorrhage (aSAH) patients and further investigated correlation of serum NOX2 levels with disease severity, delayed cerebral ischemia (DCI) plus prognosis after aSAH.Methods: Serum NOX2 levels were measured in 123 aSAH patients and 123 healthy controls. World Federation of Neurological Surgeons scale (WFNS) score and modified Fisher (mFisher) score were utilized to assess disease severity. Modified Rankin scale (mRS) score was used to evaluate the clinical prognosis at 90 days after aSAH. Relations of serum NOX2 levels to DCI and 90-day poor prognosis (mRS score of 3– 6) were analyzed using multivariate analysis. Receiver operating characteristic curve (ROC) was built to evaluate the prognostic predictive capability.Results: Serum NOX2 levels in aSAH patients, compared with healthy controls, were significantly increased, and were independently correlated with WFNS score, mFisher score and post-stroke 90-day mRS score. Patients with poor prognosis or DCI had significantly

Diffusion Models (DMs) have empowered great success in artificial-intelligence-generated content, especially in artwork creation, yet raising new concerns in intellectual properties and copyright. For example, infringers can make profits by imitating non-authorized human-created paintings with DMs. Recent researches suggest that various adversarial examples for diffusion models can be effective tools against these copyright infringements. However, current adversarial examples show weakness in transferability over different painting-imitating methods and robustness under straightforward adversarial defense, for example, noise purification. We surprisingly find that the transferability of adversarial examples can be significantly enhanced by exploiting a fused and modified adversarial loss term under consistent parameters. In this work, we comprehensively evaluate the cross-method transferability of adversarial examples. The experimental observation shows that our method generates more transferable adversarial examples with even stronger robustness against the simple adversarial defense., Comment: Working paper

While generative diffusion models excel in producing high-quality images, they can also be misused to mimic authorized images, posing a significant threat to AI systems. Efforts have been made to add calibrated perturbations to protect images from diffusion-based mimicry pipelines. However, most of the existing methods are too ineffective and even impractical to be used by individual users due to their high computation and memory requirements. In this work, we present novel findings on attacking latent diffusion models (LDM) and propose new plug-and-play strategies for more effective protection. In particular, we explore the bottleneck in attacking an LDM, discovering that the encoder module rather than the denoiser module is the vulnerable point. Based on this insight, we present our strategy using Score Distillation Sampling (SDS) to double the speed of protection and reduce memory occupation by half without compromising its strength. Additionally, we provide a robust protection strategy by counterintuitively minimizing the semantic loss, which can assist in generating more natural perturbations. Finally, we conduct extensive experiments to substantiate our findings and comprehensively evaluate our newly proposed strategies. We hope our insights and protective measures can contribute to better defense against malicious diffusion-based mimicry, advancing the development of secure AI systems. The code is available in https://github.com/xavihart/Diff-Protect, Comment: ICLR 2024 Conference Paper, the code is available in https://github.com/xavihart/Diff-Protect

Extensive system scales (i.e. thousands of GPU/TPUs) and prolonged training periods (i.e. months of pretraining) significantly escalate the probability of failures when training large language models (LLMs). Thus, efficient and reliable fault-tolerance methods are in urgent need. Checkpointing is the primary fault-tolerance method to periodically save parameter snapshots from GPU memory to disks via CPU memory. In this paper, we identify the frequency of existing checkpoint-based fault-tolerance being significantly limited by the storage I/O overheads, which results in hefty re-training costs on restarting from the nearest checkpoint. In response to this gap, we introduce an in-memory fault-tolerance framework for large-scale LLM pretraining. The framework boosts the efficiency and reliability of fault tolerance from three aspects: (1) Reduced Data Transfer and I/O: By asynchronously caching parameters, i.e., sharded model parameters, optimizer states, and RNG states, to CPU volatile memory, Our framework significantly reduces communication costs and bypasses checkpoint I/O. (2) Enhanced System Reliability: Our framework enhances parameter protection with a two-layer hierarchy: snapshot management processes (SMPs) safeguard against software failures, together with Erasure Coding (EC) protecting against node failures. This double-layered protection greatly improves the survival probability of the parameters compared to existing checkpointing methods. (3) Improved Snapshotting Frequency: Our framework achieves more frequent snapshotting compared with asynchronous checkpointing optimizations under the same saving time budget, which improves the fault tolerance efficiency. Empirical results demonstrate that Our framework minimizes the overhead of fault tolerance of LLM pretraining by effectively leveraging redundant CPU resources., Comment: Fault Tolerance, Checkpoint Optimization, Large Language Model, 3D parallelism

Adversarial attacks on Latent Diffusion Model (LDM), the state-of-the-art image generative model, have been adopted as effective protection against malicious finetuning of LDM on unauthorized images. We show that these attacks add an extra error to the score function of adversarial examples predicted by LDM. LDM finetuned on these adversarial examples learns to lower the error by a bias, from which the model is attacked and predicts the score function with biases. Based on the dynamics, we propose to improve the adversarial attack on LDM by Attacking with Consistent score-function Errors (ACE). ACE unifies the pattern of the extra error added to the predicted score function. This induces the finetuned LDM to learn the same pattern as a bias in predicting the score function. We then introduce a well-crafted pattern to improve the attack. Our method outperforms state-of-the-art methods in adversarial attacks on LDM.

Open-vocabulary detection (OVD) is a new object detection paradigm, aiming to localize and recognize unseen objects defined by an unbounded vocabulary. This is challenging since traditional detectors can only learn from pre-defined categories and thus fail to detect and localize objects out of pre-defined vocabulary. To handle the challenge, OVD leverages pre-trained cross-modal VLM, such as CLIP, ALIGN, etc. Previous works mainly focus on the open vocabulary classification part, with less attention on the localization part. We argue that for a good OVD detector, both classification and localization should be parallelly studied for the novel object categories. We show in this work that improving localization as well as cross-modal classification complement each other, and compose a good OVD detector jointly. We analyze three families of OVD methods with different design emphases. We first propose a vanilla method,i.e., cropping a bounding box obtained by a localizer and resizing it into the CLIP. We next introduce another approach, which combines a standard two-stage object detector with CLIP. A two-stage object detector includes a visual backbone, a region proposal network (RPN), and a region of interest (RoI) head. We decouple RPN and ROI head (DRR) and use RoIAlign to extract meaningful features. In this case, it avoids resizing objects. To further accelerate the training time and reduce the model parameters, we couple RPN and ROI head (CRR) as the third approach. We conduct extensive experiments on these three types of approaches in different settings. On the OVD-COCO benchmark, DRR obtains the best performance and achieves 35.8 Novel AP$_{50}$, an absolute 2.8 gain over the previous state-of-the-art (SOTA). For OVD-LVIS, DRR surpasses the previous SOTA by 1.9 AP$_{50}$ in rare categories. We also provide an object detection dataset called PID and provide a baseline on PID.

Video anomaly detection under weak supervision presents significant challenges, particularly due to the lack of frame-level annotations during training. While prior research has utilized graph convolution networks and self-attention mechanisms alongside multiple instance learning (MIL)-based classification loss to model temporal relations and learn discriminative features, these methods often employ multi-branch architectures to capture local and global dependencies separately, resulting in increased parameters and computational costs. Moreover, the coarse-grained interclass separability provided by the binary constraint of MIL-based loss neglects the fine-grained discriminability within anomalous classes. In response, this paper introduces a weakly supervised anomaly detection framework that focuses on efficient context modeling and enhanced semantic discriminability. We present a Temporal Context Aggregation (TCA) module that captures comprehensive contextual information by reusing the similarity matrix and implementing adaptive fusion. Additionally, we propose a Prompt-Enhanced Learning (PEL) module that integrates semantic priors using knowledge-based prompts to boost the discriminative capacity of context features while ensuring separability between anomaly sub-classes. Extensive experiments validate the effectiveness of our method's components, demonstrating competitive performance with reduced parameters and computational effort on three challenging benchmarks: UCF-Crime, XD-Violence, and ShanghaiTech datasets. Notably, our approach significantly improves the detection accuracy of certain anomaly sub-classes, underscoring its practical value and efficacy. Our code is available at: https://github.com/yujiangpu20/PEL4VAD., Comment: 13 pages, 9 figures

Insulin sensitivity progressively declines with age. Currently, the mechanism underlying age-associated insulin resistance remains unknown. Here, we identify membrane-bound matrix metalloproteinase 14 (MT1-MMP/MMP14) as a central regulator of insulin sensitivity during ageing. Ageing promotes MMP14 activation in insulin-sensitive tissues, which cleaves Insulin Receptor to suppress insulin signaling. MT1-MMP inhibition restores Insulin Receptor expression, improving insulin sensitivity in aged mice. The cleavage of Insulin Receptor by MT1-MMP also contributes to obesity-induced insulin resistance and inhibition of MT1-MMP activities normalizes metabolic dysfunctions in diabetic mouse models. Conversely, overexpression of MT1-MMP in the liver reduces the level of Insulin Receptor, impairing hepatic insulin sensitivity in young mice. The soluble Insulin Receptor and circulating MT1-MMP are positively correlated in plasma from aged human subjects and non-human primates. Our findings provide mechanistic insights into regulation of insulin sensitivity during physiological ageing and highlight MT1-MMP as a promising target for therapeutic avenue against diabetes.

We consider the SUBSET SUM problem and its important variants in this paper. In the SUBSET SUM problem, a (multi-)set $X$ of $n$ positive numbers and a target number $t$ are given, and the task is to find a subset of $X$ with the maximal sum that does not exceed $t$. It is well known that this problem is NP-hard and admits fully polynomial-time approximation schemes (FPTASs). In recent years, it has been shown that there does not exist an FPTAS of running time $\tilde\OO( 1/\epsilon^{2-\delta})$ for arbitrary small $\delta>0$ assuming ($\min$,+)-convolution conjecture~\cite{bringmann2021fine}. However, the lower bound can be bypassed if we relax the constraint such that the task is to find a subset of $X$ that can slightly exceed the threshold $t$ by $\epsilon$ times, and the sum of numbers within the subset is at least $1-\tilde\OO(\epsilon)$ times the optimal objective value that respects the constraint. Approximation schemes that may violate the constraint are also known as weak approximation schemes. For the SUBSET SUM problem, there is a randomized weak approximation scheme running in time $\tilde\OO(n+ 1/\epsilon^{5/3})$ [Mucha et al.'19]. For the special case where the target $t$ is half of the summation of all input numbers, weak approximation schemes are equivalent to approximation schemes that do not violate the constraint, and the best-known algorithm runs in $\tilde\OO(n+1/\epsilon^{{3}/{2}})$ time [Bringmann and Nakos'21].

The ability to choose an appropriate camera view among multiple cameras plays a vital role in TV shows delivery. But it is hard to figure out the statistical pattern and apply intelligent processing due to the lack of high-quality training data. To solve this issue, we first collect a novel benchmark on this setting with four diverse scenarios including concerts, sports games, gala shows, and contests, where each scenario contains 6 synchronized tracks recorded by different cameras. It contains 88-hour raw videos that contribute to the 14-hour edited videos. Based on this benchmark, we further propose a new approach temporal and contextual transformer that utilizes clues from historical shots and other views to make shot transition decisions and predict which view to be used. Extensive experiments show that our method outperforms existing methods on the proposed multi-camera editing benchmark., Comment: Extended Abstract of ECCV 2022 Workshop on AI for Creative Video Editing and Understanding

Video anomaly detection is recently formulated as a multiple instance learning task under weak supervision, in which each video is treated as a bag of snippets to be determined whether contains anomalies. Previous efforts mainly focus on the discrimination of the snippet itself without modeling the temporal dynamics, which refers to the variation of adjacent snippets. Therefore, we propose a Discriminative Dynamics Learning (DDL) method with two objective functions, i.e., dynamics ranking loss and dynamics alignment loss. The former aims to enlarge the score dynamics gap between positive and negative bags while the latter performs temporal alignment of the feature dynamics and score dynamics within the bag. Moreover, a Locality-aware Attention Network (LA-Net) is constructed to capture global correlations and re-calibrate the location preference across snippets, followed by a multilayer perceptron with causal convolution to obtain anomaly scores. Experimental results show that our method achieves significant improvements on two challenging benchmarks, i.e., UCF-Crime and XD-Violence., Comment: Accepted by ICME 2022

Jun Liu,1,2,* Siyu Wu,1,2,* Jiayi Ma,1,2 Chun Liu,1 Ting Dai,1 Xiaoyu Wu,1 Hongbin Zhao,1 Dong Zhou1 1Medical Research Centre, The Affiliated Changzhou No.2 People’s Hospital of Nanjing Medical University, Changzhou, 213164, People’s Republic of China; 2Dalian Medical University, Dalian, 116044, People’s Republic of China*These authors contributed equally to this workCorrespondence: Hongbin Zhao; Dong Zhou, Medical Research Centre, The Affiliated Changzhou No.2 People’s Hospital of Nanjing Medical University, Changzhou, 213164, People’s Republic of China, Tel/Fax +86 13919761032 ; Tel/Fax +86 13861298585, Email zhao761032@163.com; zhoudong1012@163.comPurpose: To enhance the osteoinductive effect of Hydroxyapatite (HA) in bone tissue engineering, this study manufactured polycaprolactone (PCL)/gelatin (GEL)/HA nanofibrous scaffolds incorporated with different ratios of attapulgite (ATP): HA (0:3, 0:0, 1:1, 2:1 and 3:0) by high-voltage electrospinning. The synergistic effect exerted by ATP and HA on bone formation was explored both in vivo and in vitro.Methods and Results: First, we determined the group composition and crystal structure of the nanosheets by Fourier transform infrared (FTIR) and X-ray diffraction (XRD) analyses. Then, the physical properties of the scaffolds, including the modulus of elasticity, porosity and water absorption were evaluated. Moreover, the surface microstructure of the nanofibrous scaffolds was captured by Scanning electron microscopy (SEM) and Transmission Electron Microscope (TEM). The biocompatibility of the fabricated scaffolds represented by cell counting kit 8 (CCK-8) and phalloidin staining was also assessed. Next, in vitro osteogenesis was evaluated. Real-time PCR, alkaline phosphatase (ALP) staining and Alizarin red S (ARS) staining results showed that the materials incorporated with HA and ATP at a ratio of 2:1 synergistically promoted more osteoblastic differentiation and extracellular minerali

Vision-language pre-training (VLP) on large-scale datasets has shown premier performance on various downstream tasks. In contrast to plenty of available benchmarks with English corpus, large-scale pre-training datasets and downstream datasets with Chinese corpus remain largely unexplored. In this work, we build a large-scale high-quality Chinese Cross-Modal Benchmark named CCMB for the research community, which contains the currently largest public pre-training dataset Zero and five human-annotated fine-tuning datasets for downstream tasks. Zero contains 250 million images paired with 750 million text descriptions, plus two of the five fine-tuning datasets are also currently the largest ones for Chinese cross-modal downstream tasks. Along with the CCMB, we also develop a VLP framework named R2D2, applying a pre-Ranking + Ranking strategy to learn powerful vision-language representations and a two-way distillation method (i.e., target-guided Distillation and feature-guided Distillation) to further enhance the learning capability. With the Zero and the R2D2 VLP framework, we achieve state-of-the-art performance on twelve downstream datasets from five broad categories of tasks including image-text retrieval, image-text matching, image caption, text-to-image generation, and zero-shot image classification. The datasets, models, and codes are available at https://github.com/yuxie11/R2D2

We study a bilevel optimization problem which is a zero-sum Stackelberg game. In this problem, there are two players, a leader and a follower, who pick items from a common set. Both the leader and the follower have their own (multi-dimensional) budgets, respectively. Each item is associated with a profit, which is the same to the leader and the follower, and will consume the leader's (follower's) budget if it is selected by the leader (follower). The leader and the follower will select items in a sequential way: First, the leader selects items within the leader's budget. Then the follower selects items from the remaining items within the follower's budget. The goal of the leader is to minimize the maximum profit that the follower can obtain. Let $s_A$ and $s_B$ be the dimension of the leader's and follower's budget, respectively. A special case of our problem is the bilevel knapsack problem studied by Caprara et al. [SIAM Journal on Optimization, 2014], where $s_A=s_B=1$. We consider the general problem and obtain an $(s_B+\epsilon)$-approximation algorithm when $s_A$ and $s_B$ are both constant. In particular, if $s_B=1$, our algorithm implies a PTAS for the bilevel knapsack problem, which is the first O(1)-approximation algorithm. We also complement our result by showing that there does not exist any $(4/3-\epsilon)$-approximation algorithm even if $s_A=1$ and $s_B=2$. We also consider a variant of our problem with resource augmentation when $s_A$ and $s_B$ are both part of the input. We obtain an O(1)-approximation algorithm with O(1)-resource augmentation, that is, we give an algorithm that returns a solution which exceeds the given leader's budget by O(1) times, and the objective value achieved by the solution is O(1) times the optimal objective value that respects the leader's budget.

Enhancing the stability of machine learning algorithms under distributional shifts is at the heart of the Out-of-Distribution (OOD) Generalization problem. Derived from causal learning, recent works of invariant learning pursue strict invariance with multiple training environments. Although intuitively reasonable, strong assumptions on the availability and quality of environments are made to learn the strict invariance property. In this work, we come up with the ``distributional stability" notion to mitigate such limitations. It quantifies the stability of prediction mechanisms among sub-populations down to a prescribed scale. Based on this, we propose the learnability assumption and derive the generalization error bound under distribution shifts. Inspired by theoretical analyses, we propose our novel stable risk minimization (SRM) algorithm to enhance the model's stability w.r.t. shifts in prediction mechanisms ($Y|X$-shifts). Experimental results are consistent with our intuition and validate the effectiveness of our algorithm. The code can be found at https://github.com/LJSthu/SRM., Comment: Accepted at International Conference on Artificial Intelligence and Statistics (AISTATS 2024)

[ES] El desarrollo de las redes sociales ha traído consigo nuevas oportunidades para la comunicación de los principales fabricantes. Huawei es una de las compañías líderes en el mundo debido a su fuerza corporativa, su valor de marca y su capacidad de comunicación. Este estudio examina la estrategia de comunicación en redes sociales de Huawei a través de Facebook, Twitter y Linkedin. El significado teórico es complementar la estrategia teórica de comunicación en las redes sociales y el significado práctico es servir de referencia a otras marcas para llevar a cabo la comunicación en redes sociales, [EN] The development of social media has brought new opportunities for brand communication. Huawei is one of the leading companies in the world in terms of its corporate strength, brand value and communication capability. This study examines Huawei's social media communication strategy through Facebook, Twitter and LinkedIn. The theoretical significance is to supplement the theoretical social media communication strategy, and the practical significance is to provide reference for other companies to carry out social media communication.

[ES] El desarrollo de las redes sociales ha traído consigo nuevas oportunidades para la comunicación de los principales fabricantes. Huawei es una de las compañías líderes en el mundo debido a su fuerza corporativa, su valor de marca y su capacidad de comunicación. Este estudio examina la estrategia de comunicación en redes sociales de Huawei a través de Facebook, Twitter y Linkedin. El significado teórico es complementar la estrategia teórica de comunicación en las redes sociales y el significado práctico es servir de referencia a otras marcas para llevar a cabo la comunicación en redes sociales, [EN] The development of social media has brought new opportunities for brand communication. Huawei is one of the leading companies in the world in terms of its corporate strength, brand value and communication capability. This study examines Huawei's social media communication strategy through Facebook, Twitter and LinkedIn. The theoretical significance is to supplement the theoretical social media communication strategy, and the practical significance is to provide reference for other companies to carry out social media communication.

[ES] El desarrollo de las redes sociales ha traído consigo nuevas oportunidades para la comunicación de los principales fabricantes. Huawei es una de las compañías líderes en el mundo debido a su fuerza corporativa, su valor de marca y su capacidad de comunicación. Este estudio examina la estrategia de comunicación en redes sociales de Huawei a través de Facebook, Twitter y Linkedin. El significado teórico es complementar la estrategia teórica de comunicación en las redes sociales y el significado práctico es servir de referencia a otras marcas para llevar a cabo la comunicación en redes sociales, [EN] The development of social media has brought new opportunities for brand communication. Huawei is one of the leading companies in the world in terms of its corporate strength, brand value and communication capability. This study examines Huawei's social media communication strategy through Facebook, Twitter and LinkedIn. The theoretical significance is to supplement the theoretical social media communication strategy, and the practical significance is to provide reference for other companies to carry out social media communication.

[ES] El desarrollo de las redes sociales ha traído consigo nuevas oportunidades para la comunicación de los principales fabricantes. Huawei es una de las compañías líderes en el mundo debido a su fuerza corporativa, su valor de marca y su capacidad de comunicación. Este estudio examina la estrategia de comunicación en redes sociales de Huawei a través de Facebook, Twitter y Linkedin. El significado teórico es complementar la estrategia teórica de comunicación en las redes sociales y el significado práctico es servir de referencia a otras marcas para llevar a cabo la comunicación en redes sociales, [EN] The development of social media has brought new opportunities for brand communication. Huawei is one of the leading companies in the world in terms of its corporate strength, brand value and communication capability. This study examines Huawei's social media communication strategy through Facebook, Twitter and LinkedIn. The theoretical significance is to supplement the theoretical social media communication strategy, and the practical significance is to provide reference for other companies to carry out social media communication.

With AlphaGo defeats top human players, reinforcement learning(RL) algorithms have gradually become the code-base of building stronger artificial intelligence(AI). The RL algorithm design firstly needs to adapt to the specific environment, so the designed environment guides the rapid and profound development of RL algorithms. However, the existing environments, which can be divided into real world games and customized toy environments, have obvious shortcomings. For real world games, it is designed for human entertainment, and too much difficult for most of RL researchers. For customized toy environments, there is no widely accepted unified evaluation standard for all RL algorithms. Therefore, we introduce the first virtual user-friendly environment framework for RL. In this framework, the environment can be easily configured to realize all kinds of RL tasks in the mainstream research. Then all the mainstream state-of-the-art(SOTA) RL algorithms can be conveniently evaluated and compared. Therefore, our contributions mainly includes the following aspects: 1.single configured environment for all classification of SOTA RL algorithms; 2.combined environment of more than one classification RL algorithms; 3.the evaluation standard for all kinds of RL algorithms. With all these efforts, a possibility for breeding an AI with capability of general competency in a variety of tasks is provided, and maybe it will open up a new chapter for AI., Comment: 19 pages,16 figures

In this paper, we propose a novel multi-variate algorithm using a triple-regression methodology to predict the airborne-pollen allergy season that can be customized for each patient in the long term. To improve the prediction accuracy, we first perform a pre-processing to integrate the historical data of pollen concentration and various inferential signals from other covariates such as the meteorological data. We then propose a novel algorithm which encompasses three-stage regressions: in Stage 1, a regression model to predict the start/end date of a airborne-pollen allergy season is trained from a feature matrix extracted from 12 time series of the covariates with a rolling window; in Stage 2, a regression model to predict the corresponding uncertainty is trained based on the feature matrix and the prediction result from Stage 1; in Stage 3, a weighted linear regression model is built upon prediction results from Stage 1 and 2. It is observed and proved that Stage 3 contributes to the improved forecasting accuracy and the reduced uncertainty of the multi-variate triple-regression algorithm. Based on different allergy sensitivity level, the triggering concentration of the pollen - the definition of the allergy season can be customized individually. In our backtesting, a mean absolute error (MAE) of 4.7 days was achieved using the algorithm. We conclude that this algorithm could be applicable in both generic and long-term forecasting problems., Comment: 4 pages, 4 figures

Augmented Reality (AR) is known to enhance user experience, however, it remains under-adopted in industry. We present an AR interaction system improving human-machine coordination in Internet of Things (IoT) and Industry 4.0 applications including manufacturing and assembly, maintenance and safety, and other highly-interactive functions. A driver of slow adoption is the computational complexity and inaccuracy in localization and rendering digital content. AR systems may render digital content close to the associated physical objects, but traditional object recognition and localization modules perform poorly when tracking texture-less objects and complex shapes, presenting a need for robust and efficient digital content rendering techniques. We propose a method of improving IoT-AR by integrating Deep Learning with AR to increase accuracy and robustness of the target object localization module, taking both color and depth images as input and outputting the target's pose parameters. Quantitative and qualitative experiments prove this system's efficacy and show potential for fusing these emerging technologies in real-world applications.

The safety of the aviation industry heavily depends on the performance of pilots. A pilot with a good cockpit culture improves safety performance. However, a lack of understanding of how a well-established aviation program impacts student pilots’ cockpit culture regarding power distance, masculinity, individualism, and uncertainty avoidance. This study aimed to investigate the training effect on Chinese student perceptions regarding power distance, masculinity, individualism, and uncertainty avoidance, and its related variables. The study collected 726 Chinese student pilots and analyzed their cockpit culture based on the survey data. The study examined the academic tenure influences on cockpit culture variables within student pilots. The profile analysis indicated that first-year students, sophomores, juniors, and seniors had different profiles regarding the cockpit culture variables. Outcomes show that there is a positive influence on Chinese student pilots’ perception of power distance and individualism. The perception of masculinity increases with their academic tenures and uncertainty avoidance has a periodical pattern.

Xiaoyu Wu,1 Facai Cui,1 Yu Chen,2 Ya Zhu,1 Fengzhen Liu1 1Department of Clinical Laboratory, Henan Provincial People’s Hospital, People’s Hospital of Zhengzhou University, Zhengzhou, Henan, People’s Republic of China; 2Department of Pathology, Affiliated Tumor Hospital of Zhengzhou University, Zhengzhou, Henan, People’s Republic of ChinaCorrespondence: Xiaoyu Wu Email xiaoyuwhn@163.comIntroduction: Colorectal cancer (CRC), the third most common cancer worldwide, involves a physiological and pathological long non-coding RNA (lncRNA) paradigm shift. It has been reported that the lncRNA LOXL1-AS1 affects tumor development for many kinds of cancers, but its functions and mechanisms in CRC remain unknown.Methods: Expression levels of LOXL1-AS1 and miR-708-5p within CRC tissues and cell lines were measured using qRT-PCR. The performance of gain-of-function and loss-of-function assays was aimed at examining the effects of LOXL1-AS1 and miR-708-5p; colony formation and cell viability assays were carried out to measure cell multiplication; and Transwell migration and wound-healing assays were carried out for the measurement of cell migration and invasion. Luciferase reporter assay was used to verify the interactions between LOXL1-AS1 and miR-708-5p and between miR-708-5p and the CD44-EGFR signaling pathway. Finally, expression of CD44 and EGFR proteins was measured by Western blot and immunofluorescence assays.Results: In this study, we reveal that the regulation of lncRNA LOXL1-AS1 occurs within CRC based on the correlation with poor clinical outcomes. LOXL1-AS1 knockdown along with miR-708-5p overpresentation in CRC cell lines inhibited cell multiplication, migration, and invasion. The inhibiting effect of LOXL1-AS1 knockdown on CRC was reversed by upregulating the CD44-EGFR signal pathway. From the perspective of mechanism, LOXL1-AS1 imposes sponging upon miR-708-5p and thereby promotes the CD44-EGFR signal pathway in CRC cells.Disc

Nanocrystalline (NC) materials have fascinating physical and chemical properties, thereby they exhibit great prospects in academic and industrial fields. Highly efficient approaches for fabricating bulk NC materials have been pursued extensively over past decades. However, the instability of nanograin, which is sensitive to processing parameters (such as temperature and time), is always a challenging issue to be solved and remains to date. Herein, we report an ultrafast nanostructuring strategy, namely ultrasonic vibration consolidation (UVC). The strategy utilizes internal friction heat, generated from mutually rubbing between Ti-based metallic glass powders, to heat the glassy alloy rapidly through its supercooled liquid regime, and accelerated viscous flow bonds the powders together. Consequently, bulk NC-Ti alloy with grain size ranging from 10 to 70 nm and nearly full density is consolidated in 2 seconds. The novel consolidation approach proposed here offers a general and highly efficient pathway for manufacturing bulk nanomaterials.

Nanocrystalline (NC) materials have fascinating physical and chemical properties, thereby they exhibit great prospects in academic and industrial fields. Highly efficient approaches for fabricating bulk NC materials have been pursued extensively over past decades. However, the instability of nanograin, which is sensitive to processing parameters (such as temperature and time), is always a challenging issue to be solved and remains to date. Herein, we report an ultrafast nanostructuring strategy, namely ultrasonic vibration consolidation (UVC). The strategy utilizes internal friction heat, generated from mutually rubbing between Ti-based metallic glass powders, to heat the glassy alloy rapidly through its supercooled liquid regime, and accelerated viscous flow bonds the powders together. Consequently, bulk NC-Ti alloy with grain size ranging from 10 to 70 nm and nearly full density is consolidated in 2 seconds. The novel consolidation approach proposed here offers a general and highly efficient pathway for manufacturing bulk nanomaterials.

This paper proposes an interactive image-segmentation method which is based on superpixel. To achieve fast segmentation, the method is used to establish a Graphcut model using superpixels as nodes, and a new energy function is proposed. Experimental results demonstrate that the authors' method has excellent performance in terms of segmentation accuracy and computation efficiency compared with other segmentation algorithm based on pixels.

This paper proposes an interactive image-segmentation method which is based on superpixel. To achieve fast segmentation, the method is used to establish a Graphcut model using superpixels as nodes, and a new energy function is proposed. Experimental results demonstrate that the authors' method has excellent performance in terms of segmentation accuracy and computation efficiency compared with other segmentation algorithm based on pixels.

Plaque accumulation in the walls of coronary arteries reduces the delivery of nutrients and oxygen to the myocardium. This luminal narrowing can cause clinical indications such as angina or heart attacks, and without treatment, can be fatal. One method of treatment is the percutaneous intervention of stents to re-canalize the vessel. A potential complication of stent implantation is arterial wall remodeling and renarrowing of the vessel; termed restenosis. This can be prevented in the majority of patients with an antiproliferative drug coating on the surface of the stent: a drug-eluting stent. I hypothesize that drug transport in the arterial wall from these devices varies between arterial locations (left anterior descending (LADC) versus right (RC) coronary artery; proximal, middle, versus distal regions). The purpose of this work was to identify the properties of the vascular wall that govern transport, and computationally model stent-based delivery to better understand any differences that could exist in transport based on location. The first aim of this work was to identify the porohyperelastic properties. Permeability showed a decrease along the length of the LADC artery of 198%, and 98.6% along the length of the RC artery (p=NS between LADC and RC). Mechanical properties indicated significant differences between the LADC and RC arteries, with the LADC artery being stiffer than the RC. The second aim of this work was to identify the mass transport and cellular binding properties. There was no difference between the LADC and RC arteries; however, diffusivity peaked in the middle region of both arteries by a factor of 2.07. Convection coupling coefficients indicated an upward trend down each artery with the RC artery having higher values. The third aim was to use the model constants from the previous two aims to create six parametric computational models of stent deployment and drug delivery into the respective arterial sections. Results indicated that RC section

China, the most populous country in the world, has had high economic growth during the past two decades. While economic reforms in China have received a great deal of attention, researchers have paid far less attention to the effects of population policies. In this dissertation, I discuss the One-Child policy in China and its impact on human capital accumulation. I examine the impact of family size on parental health outcomes by exploiting the exogenous change in family size under the One-Child Policy in China. The results indicate that the number of children in a family significantly affects measures of health, such as weight and blood pressure. The impact of women's relative bargaining power on household food consumption and health outcomes are also examined in this dissertation. I find no significant effects of bargaining power and common preference could not be rejected.

Chen, P., Liao, W. B., Liu, L. H., Luo, F., Wu, X. Y., Li, P. J., . . . Liu, Z. Y. (2018). Ultrafast consolidation of bulk nanocrystalline titanium alloy through ultrasonic vibration. Scientific Reports, 8(1). Available here