Search

Design activity -- constructing an artifact description satisfying given goals and constraints -- distinguishes humanity from other animals and traditional machines, and endowing machines with design abilities at the human level or beyond has been a long-term pursuit. Though machines have already demonstrated their abilities in designing new materials, proteins, and computer programs with advanced artificial intelligence (AI) techniques, the search space for designing such objects is relatively small, and thus, "Can machines design like humans?" remains an open question. To explore the boundary of machine design, here we present a new AI approach to automatically design a central processing unit (CPU), the brain of a computer, and one of the world's most intricate devices humanity have ever designed. This approach generates the circuit logic, which is represented by a graph structure called Binary Speculation Diagram (BSD), of the CPU design from only external input-output observations instead of formal program code. During the generation of BSD, Monte Carlo-based expansion and the distance of Boolean functions are used to guarantee accuracy and efficiency, respectively. By efficiently exploring a search space of unprecedented size 10^{10^{540}}, which is the largest one of all machine-designed objects to our best knowledge, and thus pushing the limits of machine design, our approach generates an industrial-scale RISC-V CPU within only 5 hours. The taped-out CPU successfully runs the Linux operating system and performs comparably against the human-designed Intel 80486SX CPU. In addition to learning the world's first CPU only from input-output observations, which may reform the semiconductor industry by significantly reducing the design cycle, our approach even autonomously discovers human knowledge of the von Neumann architecture., 28 pages

Domain adaptation in reinforcement learning (RL) mainly deals with the changes of observation when transferring the policy to a new environment. Many traditional approaches of domain adaptation in RL manage to learn a mapping function between the source and target domain in explicit or implicit ways. However, they typically require access to abundant data from the target domain. Besides, they often rely on visual clues to learn the mapping function and may fail when the source domain looks quite different from the target domain. To address these problems, we propose a novel framework Online Prototype Alignment (OPA) to learn the mapping function based on the functional similarity of elements and is able to achieve the few-shot policy transfer within only several episodes. The key insight of OPA is to introduce an exploration mechanism that can interact with the unseen elements of the target domain in an efficient and purposeful manner, and then connect them with the seen elements in the source domain according to their functionalities (instead of visual clues). Experimental results show that when the target domain looks visually different from the source domain, OPA can achieve better transfer performance even with much fewer samples from the target domain, outperforming prior methods., This paper has been accepted at ICML2023

Unlearning techniques are proposed to prevent third parties from exploiting unauthorized data, which generate unlearnable samples by adding imperceptible perturbations to data for public publishing. These unlearnable samples effectively misguide model training to learn perturbation features but ignore image semantic features. We make the in-depth analysis and observe that models can learn both image features and perturbation features of unlearnable samples at an early stage, but rapidly go to the overfitting stage since the shallow layers tend to overfit on perturbation features and make models fall into overfitting quickly. Based on the observations, we propose Progressive Staged Training to effectively prevent models from overfitting in learning perturbation features. We evaluated our method on multiple model architectures over diverse datasets, e.g., CIFAR-10, CIFAR-100, and ImageNet-mini. Our method circumvents the unlearnability of all state-of-the-art methods in the literature and provides a reliable baseline for further evaluation of unlearnable techniques.

ObjectiveThere are few studies on the prevalence and factors associated with frailty and pre-frailty in older adults with asthma worldwide. The aim of this study was to examine the epidemiological status and factors associated with frailty and pre-frailty in older adults with asthma in China.Research design and methodsData were obtained from the Sample Survey of Aged Population in Urban and Rural China in 2015, a nationwide cross-sectional survey covering 224,142 older people aged 60 years or older in 31 provinces/autonomous regions/municipalities in mainland China. We performed frailty and pre-frailty assessments using the frailty index, and the diagnosis of asthma in the older adults was self-reported based on the history of the physician's diagnosis.ResultsNine thousand four hundred sixteen older adults with asthma were included in the study. The age-sex standardized prevalence of frailty and pre-frailty in Chinese older adults with asthma was 35.8% (95% CI 34.8%−36.7%) and 54.5% (95% CI 53.5%−55.5%). Multinomial logistic regression analysis showed that increased age, female, illiteracy, living alone, poor economic status, ADL disability, comorbid chronic diseases, previous hospitalization in the past year, and residence in northern China were associated with frailty and pre-frailty in older adults with asthma.ConclusionThe prevalence of frailty and pre-frailty in Chinese older adults with asthma is very high, and assessment of frailty should become routine in the management of older adults with asthma. Appropriate public health prevention strategies based on identified risk factors for frailty in older adults with asthma should be developed to reduce the burden of frailty in Chinese older adults with asthma.

The training for deep neural networks (DNNs) demands immense energy consumption, which restricts the development of deep learning as well as increases carbon emissions. Thus, the study of energy-efficient training for DNNs is essential. In training, the linear layers consume the most energy because of the intense use of energy-consuming full-precision (FP32) multiplication in multiply-accumulate (MAC). The energy-efficient works try to decrease the precision of multiplication or replace the multiplication with energy-efficient operations such as addition or bitwise shift, to reduce the energy consumption of FP32 multiplications. However, the existing energy-efficient works cannot replace all of the FP32 multiplications during both forward and backward propagation with low-precision energy-efficient operations. In this work, we propose an Adaptive Layer-wise Scaling PoT Quantization (ALS-POTQ) method and a Multiplication-Free MAC (MF-MAC) to replace all of the FP32 multiplications with the INT4 additions and 1-bit XOR operations. In addition, we propose Weight Bias Correction and Parameterized Ratio Clipping techniques for stable training and improving accuracy. In our training scheme, all of the above methods do not introduce extra multiplications, so we reduce up to 95.8% of the energy consumption in linear layers during training. Experimentally, we achieve an accuracy degradation of less than 1% for CNN models on ImageNet and Transformer model on the WMT En-De task. In summary, we significantly outperform the existing methods for both energy efficiency and accuracy.

Despite the broad application of deep reinforcement learning (RL), transferring and adapting the policy to unseen but similar environments is still a significant challenge. Recently, the language-conditioned policy is proposed to facilitate policy transfer through learning the joint representation of observation and text that catches the compact and invariant information across environments. Existing studies of language-conditioned RL methods often learn the joint representation as a simple latent layer for the given instances (episode-specific observation and text), which inevitably includes noisy or irrelevant information and cause spurious correlations that are dependent on instances, thus hurting generalization performance and training efficiency. To address this issue, we propose a conceptual reinforcement learning (CRL) framework to learn the concept-like joint representation for language-conditioned policy. The key insight is that concepts are compact and invariant representations in human cognition through extracting similarities from numerous instances in real-world. In CRL, we propose a multi-level attention encoder and two mutual information constraints for learning compact and invariant concepts. Verified in two challenging environments, RTFM and Messenger, CRL significantly improves the training efficiency (up to 70%) and generalization ability (up to 30%) to the new environment dynamics., Comment: Accepted by AAAI 2023

Additional file 2: Figure S2. The heatmap of weekly positive rate of IFV among children with URTI and LRTI in different diagnosis, gender and age groups during January 2014 to June 2022. Weekly positive rate of IFV−A and IFV−B among children with URTI;Weekly positive rate of IFV among children with URTI in different gender;Weekly positive rate of IFV among children with URTI in different age groups;Weekly positive rate of IFV−A and IFV−B among children with LRTI;Weekly positive rate of IFV among children with LRTI in different gender;Weekly positive rate of IFV among children with LRTI in different age groups.

Additional file 1: Figure S1. The positive rate of IFV among children with URTI and LRTI in different gender and age groups during January 2014 to June 2022.The positive rate of IFV−A and IFV−B among children with URTI;The positive rate of IFV among children with URTI in different age groups;The positive rate of IFV among children with URTI in different gender;The positive rate of IFV−A and IFV−B among children with LRTI;The positive rate of IFV among children with LRTI in different age groups;The positive rate of IFV among children with LRTI in different gender.

Symbolic regression, the task of extracting mathematical expressions from the observed data $\{ \vx_i, y_i \}$, plays a crucial role in scientific discovery. Despite the promising performance of existing methods, most of them conduct symbolic regression in an \textit{offline} setting. That is, they treat the observed data points as given ones that are simply sampled from uniform distributions without exploring the expressive potential of data. However, for real-world scientific problems, the data used for symbolic regression are usually actively obtained by doing experiments, which is an \textit{online} setting. Thus, how to obtain informative data that can facilitate the symbolic regression process is an important problem that remains challenging. In this paper, we propose QUOSR, a \textbf{qu}ery-based framework for \textbf{o}nline \textbf{s}ymbolic \textbf{r}egression that can automatically obtain informative data in an iterative manner. Specifically, at each step, QUOSR receives historical data points, generates new $\vx$, and then queries the symbolic expression to get the corresponding $y$, where the $(\vx, y)$ serves as new data points. This process repeats until the maximum number of query steps is reached. To make the generated data points informative, we implement the framework with a neural network and train it by maximizing the mutual information between generated data points and the target expression. Through comprehensive experiments, we show that QUOSR can facilitate modern symbolic regression methods by generating informative data.

Additional file 3: Table S1. The positive rate of IFV in Hubei Province from 2017-2022, stratified by COVID-19 outbreak.

The advents of Large Language Models (LLMs) have shown promise in augmenting programming using natural interactions. However, while LLMs are proficient in compiling common usage patterns into a programming language, e.g., Python, it remains a challenge how to edit and debug an LLM-generated program. We introduce ANPL, a programming system that allows users to decompose user-specific tasks. In an ANPL program, a user can directly manipulate sketch, which specifies the data flow of the generated program. The user annotates the modules, or hole with natural language descriptions offloading the expensive task of generating functionalities to the LLM. Given an ANPL program, the ANPL compiler generates a cohesive Python program that implements the functionalities in hole, while respecting the dataflows specified in sketch. We deploy ANPL on the Abstraction and Reasoning Corpus (ARC), a set of unique tasks that are challenging for state-of-the-art AI systems, showing it outperforms baseline programming systems that (a) without the ability to decompose tasks interactively and (b) without the guarantee that the modules can be correctly composed together. We obtain a dataset consisting of 300/400 ARC tasks that were successfully decomposed and grounded in Python, providing valuable insights into how humans decompose programmatic tasks. See the dataset at https://iprc-dip.github.io/DARC.

Diffusion models have demonstrated remarkable performance in image generation tasks, paving the way for powerful AIGC applications. However, these widely-used generative models can also raise security and privacy concerns, such as copyright infringement, and sensitive data leakage. To tackle these issues, we propose a method, Unlearnable Diffusion Perturbation, to safeguard images from unauthorized exploitation. Our approach involves designing an algorithm to generate sample-wise perturbation noise for each image to be protected. This imperceptible protective noise makes the data almost unlearnable for diffusion models, i.e., diffusion models trained or fine-tuned on the protected data cannot generate high-quality and diverse images related to the protected training data. Theoretically, we frame this as a max-min optimization problem and introduce EUDP, a noise scheduler-based method to enhance the effectiveness of the protective noise. We evaluate our methods on both Denoising Diffusion Probabilistic Model and Latent Diffusion Models, demonstrating that training diffusion models on the protected data lead to a significant reduction in the quality of the generated images. Especially, the experimental results on Stable Diffusion demonstrate that our method effectively safeguards images from being used to train Diffusion Models in various tasks, such as training specific objects and styles. This achievement holds significant importance in real-world scenarios, as it contributes to the protection of privacy and copyright against AI-generated content.

Recently, deep learning techniques have shown great success in automatic code generation. Inspired by the code reuse, some researchers propose copy-based approaches that can copy the content from similar code snippets to obtain better performance. Practically, human developers recognize the content in the similar code that is relevant to their needs, which can be viewed as a code sketch. The sketch is further edited to the desired code. However, existing copy-based approaches ignore the code sketches and tend to repeat the similar code without necessary modifications, which leads to generating wrong results. In this paper, we propose a sketch-based code generation approach named SkCoder to mimic developers' code reuse behavior. Given a natural language requirement, SkCoder retrieves a similar code snippet, extracts relevant parts as a code sketch, and edits the sketch into the desired code. Our motivations are that the extracted sketch provides a well-formed pattern for telling models "how to write". The post-editing further adds requirement-specific details to the sketch and outputs the complete code. We conduct experiments on two public datasets and a new dataset collected by this work. We compare our approach to 20 baselines using 5 widely used metrics. Experimental results show that (1) SkCoder can generate more correct programs, and outperforms the state-of-the-art - CodeT5-base by 30.30%, 35.39%, and 29.62% on three datasets. (2) Our approach is effective to multiple code generation models and improves them by up to 120.1% in Pass@1. (3) We investigate three plausible code sketches and discuss the importance of sketches. (4) We manually evaluate the generated code and prove the superiority of our SkCoder in three aspects., Comment: Accepted by the 45th IEEE/ACM International Conference on Software Engineering (ICSE 2023)

To observe the effect of electroacupuncture (EA) on the ocular surface inflammation and α7 nicotinic acetylcholine receptor (α7nAChR) / nuclear factor kappa-B (NF-κB) p65 signal pathway in guinea pigs with dry eye, so as to explore its underlying mechanism.A total of 32 male British tricolor short haired guinea pigs were randomized into blank control, model, EA and sham acupuncture groups, with 8 guinea pigs in each group. The dry eye model was established by subcutaneous injection of scopolamine hydrobromide solution (0.6 mg/0.2 mL each time, 4 times a day for 10 days). Guinea pigs of the EA group was treated with EA at bilateral "Cuanzhu" (BL2) and "Taiyang" (HN5), and manual acupuncture at bilateral "Jingming" (BL1), "Sizhukong" (SJ23), "Tongziliao" (GB1) for 15 min, once daily for 14 days. For animals of the sham acupuncture group, a blunt needle was used to prick the skin surface of the acupoints, the acupoint selection and stimulation time were the same as those in the EA group. Before and after modeling and after the intervention, the breakup time (BUT) of lacrimal film, sodium fluorescein coloring (Fl) state of corneal epithelium and phenol red thread (PRT) moist length were recorded for assessing the severity of dry eye. The density of activated immune cells around the corneal epithelial stromal cells was determined by corneal confocal microscopy. The contents of interleukin-4 (IL-4), IL-6, IL-10, tumor necrosis factor α (TNF-α) in the cornea and lacri-mal gland tissues were determined by ELISA, and the expression levels of α7nAChR and NF-κB p65 in the cornea and lacrimal gland were detected by immunohistochemistry and Western blot, separately.Compared with the blank control group, the corneal Fl, density of activated immune cells of corneal epithelium, contents of IL-6, IL-10 and TNF-α in both corneal and lacrimal gland tissues, NF-κB p65 cell positive rate and protein expression of lacrimal gland and corneal tissues were significantly increased (EA can improve corneal and lacrimal function in dry eye guinea pigs, which may be associated with its actions in increasing the expression of α7nAChR, inhibiting the nuclear translocation of NF-κB, and reducing the activated immune cells and inflammatory reaction.

In the software engineering community, deep learning (DL) has recently been applied to many source code processing tasks. Due to the poor interpretability of DL models, their security vulnerabilities require scrutiny. Recently, researchers have identified an emergent security threat, namely poison attack. The attackers aim to inject insidious backdoors into models by poisoning the training data with poison samples. Poisoned models work normally with clean inputs but produce targeted erroneous results with poisoned inputs embedded with triggers. By activating backdoors, attackers can manipulate the poisoned models in security-related scenarios. To verify the vulnerability of existing deep source code processing models to the poison attack, we present a poison attack framework for source code named CodePoisoner as a strong imaginary enemy. CodePoisoner can produce compilable even human-imperceptible poison samples and attack models by poisoning the training data with poison samples. To defend against the poison attack, we further propose an effective defense approach named CodeDetector to detect poison samples in the training data. CodeDetector can be applied to many model architectures and effectively defend against multiple poison attack approaches. We apply our CodePoisoner and CodeDetector to three tasks, including defect detection, clone detection, and code repair. The results show that (1) CodePoisoner achieves a high attack success rate (max: 100%) in misleading models to targeted erroneous behaviors. It validates that existing deep source code processing models have a strong vulnerability to the poison attack. (2) CodeDetector effectively defends against multiple poison attack approaches by detecting (max: 100%) poison samples in the training data. We hope this work can help practitioners notice the poison attack and inspire the design of more advanced defense techniques., 23 pages, 9 figures

Object-oriented reinforcement learning (OORL) is a promising way to improve the sample efficiency and generalization ability over standard RL. Recent works that try to solve OORL tasks without additional feature engineering mainly focus on learning the object representations and then solving tasks via reasoning based on these object representations. However, none of these works tries to explicitly model the inherent similarity between different object instances of the same category. Objects of the same category should share similar functionalities; therefore, the category is the most critical property of an object. Following this insight, we propose a novel framework named Object-Category Aware Reinforcement Learning (OCARL), which utilizes the category information of objects to facilitate both perception and reasoning. OCARL consists of three parts: (1) Category-Aware Unsupervised Object Discovery (UOD), which discovers the objects as well as their corresponding categories; (2) Object-Category Aware Perception, which encodes the category information and is also robust to the incompleteness of (1) at the same time; (3) Object-Centric Modular Reasoning, which adopts multiple independent and object-category-specific networks when reasoning based on objects. Our experiments show that OCARL can improve both the sample efficiency and generalization in the OORL domain., This paper is to be published on NeurIPS 2022

The deep neural network (DNN) models are deemed confidential due to their unique value in expensive training efforts, privacy-sensitive training data, and proprietary network characteristics. Consequently, the model value raises incentive for adversary to steal the model for profits, such as the representative model extraction attack. Emerging attack can leverage timing-sensitive architecture-level events (i.e., Arch-hints) disclosed in hardware platforms to extract DNN model layer information accurately. In this paper, we take the first step to uncover the root cause of such Arch-hints and summarize the principles to identify them. We then apply these principles to emerging Unified Memory (UM) management system and identify three new Arch-hints caused by UM's unique data movement patterns. We then develop a new extraction attack, UMProbe. We also create the first DNN benchmark suite in UM and utilize the benchmark suite to evaluate UMProbe. Our evaluation shows that UMProbe can extract the layer sequence with an accuracy of 95% for almost all victim test models, which thus calls for more attention to the DNN security in UM system.

In more than 30 years of aptamer research, it has become widely accepted that aptamers are fascinating binding molecules for a vast variety of applications. However, the majority of targets have been proteins, although special variants of the so-called SELEX process for the molecular evolution of specific aptamers have also been developed, allowing for the targeting of small molecules as well as larger structures such as cells and even cellular networks of human (tumor) tissues. Although the provocative thesis is widely accepted in the field, that is, in principle, any level of complexity for SELEX targets is possible, the number of studies on whole organs or at least parts of them is limited. To pioneer this thesis, and based on our FluCell-SELEX process, here, we have developed polyclonal aptamer libraries against apices and the elongation/differentiation zones of plant roots as examples of organs. We show that dedicated libraries can specifically label the respective parts of the root, allowing us to distinguish them in fluorescence microscopy. We consider this achievement to be an initial but important evidence for the robustness of this SELEX variant. These libraries may be valuable tools for plant research and a promising starting point for the isolation of more specific individual aptamers directed against root-specific epitopes.

Recent studies have demonstrated that changes in the abundance of the intestinal bacterium

Easy and reliable identification of pathogenic species such as yeasts, emerging as problematic microbes originating from the genus

Dry eye is a common ocular surface disease accompanied by ocular discomfort, which severely affects the work and life of patients. There are numerous causes of dry eye, and the disease is easy to recur, which is difficult to cure effectively. With the application of polymerase chain reaction(PCR)and western blotting in study of dry eye, genomics events such as metagenomics, genetically engineered animals and gene transfection have been further explored in dry eye research. Advances in the application of genomics for dry eye are summarized, in order to provide an important basis for further revealing the risk factors and pathogenesis, and exploring new drugs and therapies for effective treatment of dry eye.

Background: Extracellular vesicles (EVs) were reported to participate in various cellular processes based on the biomolecules, particularly microRNAs. Numerous commercial EVs isolation reagents are available. However, whether these reagents are suitable for separating EVs from the culture medium supernatant supernatant of model cell lines, such as HepG2, and whether the isolated products are suitable for High-throughput sequencing remains unclear.Methods: We examined three commonly used EVs isolation kits: the ExoQuick-TC exosome precipitation solution (EQ), Total Exosome Isolation from cell culture medium (EI), and exoEasy Maxi Kit (EM), to isolate EVs from HepG2 cell culture medium supernatants. EVs were identified based on marker proteins, particle size measurements, and electron microscopy analysis. The total amounts of microRNA and microRNA High-throughput sequencing data quality from EVs isolated by each kit were compared.Results: The total amount of EVs’ microRNA isolated from the EI and EM groups were higher than that obtained from the EQ group (EQ/EI: p = 0.036, EI/EM: p = 0.024). High-throughput sequencing data quality evaluation showed that the EI group possessed higher quality than those in the EM group.Conclusion: For the cell culture medium from HepG2, EVs’ microRNA isolated by EI reagents might be more suitable for High-throughput sequencing applications.

Oligonucleotide DNA aptamers represent an emergently important class of binding entities towards as different analytes as small molecules or even whole cells. Without requiring the canonical isolation of individual aptamers following the SELEX process, the focused polyclonal libraries prepared by this

Background Rhizosphere fungi and endophytic fungi play key roles in plant growth and development; however, their role in the growth of Epimedium koreanum Nakai at different stages remains unclear. Here, we used the Illumina MiSeq system, a high-throughput sequencing technology, to study the endophytic fungi and rhizosphere microbiome of Korean Epimedium. Results Epimedium koreanum Nakai rhizosphere soil and leaves had highly diverse fungal communities during the growth process. The relative abundance of soil fungi in the rhizosphere stage was higher than that of leaf endophytic fungi in the early growth stage, but the overall abundance was basically equal. Sebacina is a significantly divergent fungal genera, and Sebacina sp. are present among leaf fungi species in the rhizosphere soil of Epimedium koreanum Nakai. Sebacina sp. can move to each other in rhizosphere soil fungi and leaf endophytes. VIF (variance inflation factor) analysis showed that soluble salt, whole nitrogen, alkaline lysis nitrogen, whole phosphorus, total potassium, and fast-acting potassium are useful environmental factors for rhizosphere soil and leaf endophytic fungi: potassium, total nitrogen, whole phosphorus, and three environmental factors were significantly and positively associated with the relative abundance of Sebacina sp. Conclusions (1) This study is the first to clarify the species diversity of fungi in Epimedium koreanum Nakai leaf and rhizosphere soil. (2) Different fungal communities of rhizosphere soil fungi and leaf endophytic fungi at different growth stages of Epimedium koreanum Nakai were examined. (3) Sebacina sp. can move to each other between rhizosphere soil fungi and leaf endophytic fungi. (4) Nitrogen, phosphorus and potassium elements in the environment have a significant positive effect on the relative abundance of Sebacina sp.

Autonomous driving faces great safety challenges for a lack of global perspective and the limitation of long-range perception capabilities. It has been widely agreed that vehicle-infrastructure cooperation is required to achieve Level 5 autonomy. However, there is still NO dataset from real scenarios available for computer vision researchers to work on vehicle-infrastructure cooperation-related problems. To accelerate computer vision research and innovation for Vehicle-Infrastructure Cooperative Autonomous Driving (VICAD), we release DAIR-V2X Dataset, which is the first large-scale, multi-modality, multi-view dataset from real scenarios for VICAD. DAIR-V2X comprises 71254 LiDAR frames and 71254 Camera frames, and all frames are captured from real scenes with 3D annotations. The Vehicle-Infrastructure Cooperative 3D Object Detection problem (VIC3D) is introduced, formulating the problem of collaboratively locating and identifying 3D objects using sensory inputs from both vehicle and infrastructure. In addition to solving traditional 3D object detection problems, the solution of VIC3D needs to consider the temporal asynchrony problem between vehicle and infrastructure sensors and the data transmission cost between them. Furthermore, we propose Time Compensation Late Fusion (TCLF), a late fusion framework for the VIC3D task as a benchmark based on DAIR-V2X. Find data, code, and more up-to-date information at https://thudair.baai.ac.cn/index and https://github.com/AIR-THU/DAIR-V2X., CVPR2022

To observe the effect of electroacupuncture (EA) on ocular surface sensory neuralgia and the expression of P2XMale British tricolor short haired guinea pigs were randomly divided into control, model, medication (pranoprofen), EA and sham acupuncture groups, with 8 guinea pigs in each group. The dry eye model was induced by subcutaneous injection of scopolamine hydrobromide solution (0.6 mg/0.2 mL,once daily) for 10 d. Guinea pigs in the medication group were treated by applying pranoprofen eye drops to eyes, 1 drop for one eye each time, three times a day. Guinea pigs of the EA group received EA stimulation (4 Hz/20 Hz,1 mA) of bilateral "Cuanzhu" (BL2) and "Taiyang" (HN5) and acupuncture at "Jingming" (BL1) "Sizhukong" (TE23), "Tongziliao" (GB1) for 15 min, once a day. Guinea pigs in the sham acupuncture group received blunt stimu-lation at the surface of the same acupoint with the tip of the acupuncture needle, once a day. All the treatments were conducted for 14 d. The corneal epithelium fluorescein staining score (0-3 points) was given according to the number of fluorescence-positive dots and flake-like coloration, the corneal mechanical perception thread (CMPT) detected using a corneal perception meter, and the palpebral fissure height measured. The number of sensory neurons in the cornea and TG was determined by using cholera toxin subunit B conjugated with Alexa Fluor 488 fluorescence labelling, and the expression levels of P2X3R and PKC in the cornea and TG detected by using immunohistochemistry and Western blot, separately.Compared with the control group, the corneal fluorescein staining score, immunoactivity and expression of P2X3R proteins in both cornea and TG, PKC proteins in TG were significantly increased (EA can alleviate the damage of corneal epithelium and sensory neurons in dry eye model guinea pigs, which may be related to its functions in down-regu-lating the expression of P2X3R and PKC in the cornea and TG.

Aiming to find a program satisfying the user intent given input-output examples, program synthesis has attracted increasing interest in the area of machine learning. Despite the promising performance of existing methods, most of their success comes from the privileged information of well-designed input-output examples. However, providing such input-output examples is unrealistic because it requires the users to have the ability to describe the underlying program with a few input-output examples under the training distribution. In this work, we propose a query-based framework that trains a query neural network to generate informative input-output examples automatically and interactively from a large query space. The quality of the query depends on the amount of the mutual information between the query and the corresponding program, which can guide the optimization of the query framework. To estimate the mutual information more accurately, we introduce the functional space (F-space) which models the relevance between the input-output examples and the programs in a differentiable way. We evaluate the effectiveness and generalization of the proposed query-based framework on the Karel task and the list processing task. Experimental results show that the query-based framework can generate informative input-output examples which achieve and even outperform well-designed input-output examples., Accepted by ICLR 2022

As spiking neural networks (SNNs) are deployed increasingly in real-world efficiency critical applications, the security concerns in SNNs attract more attention. Currently, researchers have already demonstrated an SNN can be attacked with adversarial examples. How to build a robust SNN becomes an urgent issue. Recently, many studies apply certified training in artificial neural networks (ANNs), which can improve the robustness of an NN model promisely. However, existing certifications cannot transfer to SNNs directly because of the distinct neuron behavior and input formats for SNNs. In this work, we first design S-IBP and S-CROWN that tackle the non-linear functions in SNNs' neuron modeling. Then, we formalize the boundaries for both digital and spike inputs. Finally, we demonstrate the efficiency of our proposed robust training method in different datasets and model architectures. Based on our experiment, we can achieve a maximum $37.7\%$ attack error reduction with $3.7\%$ original accuracy loss. To the best of our knowledge, this is the first analysis on robust training of SNNs.

We first prove the equivalence of two definitions of Riemann-Liouville fractional integral on time scales, then by the concept of fractional derivative of Riemann-Liouville on time scales, we introduce fractional Sobolev spaces, characterize them, define weak fractional derivatives, and show that they coincide with the Riemann-Liouville ones on time scales. Next, we prove equivalence of some norms in the introduced spaces and derive their completeness, reflexivity, separability and some imbeddings. Finally, as an application, by constructing an appropriate variational setting, using the mountain pass theorem and the genus properties, the existence of weak solutions for a class of Kirchhoff-type fractional p-Laplacian systems on time scales with boundary condition is studied, and three results of the existence of weak solutions for this problem is obtained., arXiv admin note: substantial text overlap with arXiv:2112.05757

Long Document retrieval (DR) has always been a tremendous challenge for reading comprehension and information retrieval. The pre-training model has achieved good results in the retrieval stage and Ranking for long documents in recent years. However, there is still some crucial problem in long document ranking, such as data label noises, long document representations, negative data Unbalanced sampling, etc. To eliminate the noise of labeled data and to be able to sample the long documents in the search reasonably negatively, we propose the bag sampling method and the group-wise Localized Contrastive Estimation(LCE) method. We use the head middle tail passage for the long document to encode the long document, and in the retrieval, stage Use dense retrieval to generate the candidate's data. The retrieval data is divided into multiple bags at the ranking stage, and negative samples are selected in each bag. After sampling, two losses are combined. The first loss is LCE. To fit bag sampling well, after query and document are encoded, the global features of each group are extracted by convolutional layer and max-pooling to improve the model's resistance to the impact of labeling noise, finally, calculate the LCE group-wise loss. Notably, our model shows excellent performance on the MS MARCO Long document ranking leaderboard.

Canine models are increasingly being used in metabolic studies due to their physiological similarity with humans. The present study aimed to identify changes in metabolic pathways and biomarkers with potential clinical utility in a canine model of obesity and metabolic disorders induced by a high-fat diet (HFD). Eighteen male beagles were included in this study, 9 of which were fed a HFD for 24 weeks, and the remaining 9 were fed normal chow (NC) during the same period. Plasma and urine samples were collected at weeks 12 and 24 for untargeted metabolomic analysis. Dogs fed a HFD showed a gradual body weight increase during the feeding period and had hyperlipidemia, increased leukocyte counts, and impaired insulin sensitivity at week 24. Plasma and urine metabonomics analysis displayed clear separations between the HFD-fed and NC-fed dogs. A total of 263 plasma metabolites varied between the two groups, including stearidonic acid, linolenic acid, carnitine, long-chain ceramide, 3-methylxanthine, and theophylline, which are mainly engaged in fatty acid metabolism, sphingolipid metabolism, and caffeine metabolism. A total of 132 urine metabolites related to HFD-induced obesity and metabolic disorders were identified, including 3-methylxanthine, theophylline, pyridoxal 5’-phosphate, and harmine, which participate in pathways such as caffeine metabolism and vitamin digestion and absorption. Eight metabolites with increased abundance (e.g., 3-methylxanthine, theophylline, and harmine) and 4 metabolites with decreased abundance (e.g., trigonelline) in both the plasma and urine of the HFD-fed dogs were identified. In conclusion, the metabolomic analysis revealed molecular events underlying a canine HFD model and identified several metabolites as potential targets for the prevention and treatment of obesity-related metabolic disorders.

DNN-based video object detection (VOD) powers autonomous driving and video surveillance industries with rising importance and promising opportunities. However, adversarial patch attack yields huge concern in live vision tasks because of its practicality, feasibility, and powerful attack effectiveness. This work proposes Themis, a software/hardware system to defend against adversarial patches for real-time robust video object detection. We observe that adversarial patches exhibit extremely localized superficial feature importance in a small region with non-robust predictions, and thus propose the adversarial region detection algorithm for adversarial effect elimination. Themis also proposes a systematic design to efficiently support the algorithm by eliminating redundant computations and memory traffics. Experimental results show that the proposed methodology can effectively recover the system from the adversarial attack with negligible hardware overhead.

Hierarchical reinforcement learning (HRL) effectively improves agents' exploration efficiency on tasks with sparse reward, with the guide of high-quality hierarchical structures (e.g., subgoals or options). However, how to automatically discover high-quality hierarchical structures is still a great challenge. Previous HRL methods can hardly discover the hierarchical structures in complex environments due to the low exploration efficiency by exploiting the randomness-driven exploration paradigm. To address this issue, we propose CDHRL, a causality-driven hierarchical reinforcement learning framework, leveraging a causality-driven discovery instead of a randomness-driven exploration to effectively build high-quality hierarchical structures in complicated environments. The key insight is that the causalities among environment variables are naturally fit for modeling reachable subgoals and their dependencies and can perfectly guide to build high-quality hierarchical structures. The results in two complex environments, 2D-Minecraft and Eden, show that CDHRL significantly boosts exploration efficiency with the causality-driven paradigm., Comment: Accepted by NeurIPS 2022

Developers often perform repetitive code editing activities for various reasons (e.g., code refactor) during software development. Many deep learning models are applied to automate code editing by learning from the code editing history. Recently, pre-trained code editing models have achieved the state-of-the-art (SOTA) results. Pre-trained models are first pre-trained with pre-training tasks and fine-tuned with the code editing task. Existing pre-training tasks mainly are code infilling tasks (e.g., masked language modeling), which are derived from the natural language processing field and are not designed for code editing. In this paper, we propose a pre-training task specialized in code editing and present an effective pre-trained code editing model named CodeEditor. Our pre-training task further improves the performance and generalization ability of code editing models. Specifically, we collect real-world code snippets as the ground truth and use a generator to rewrite them into natural but inferior versions. Then, we pre-train our CodeEditor to edit inferior versions into the ground truth, to learn edit patterns. We conduct experiments on four datasets and evaluate models in three settings. (1) In the fine-tuning setting, we fine-tune the pre-trained CodeEditor with four datasets. CodeEditor outperforms SOTA baselines by 15%, 25.5%, and 9.4% and 26.6% on four datasets. (2) In the few-shot setting, we fine-tune the pre-trained CodeEditor with limited data. CodeEditor substantially performs better than all baselines, even outperforming baselines that are fine-tuned with all data. (3) In the zero-shot setting, we evaluate the pre-trained CodeEditor without fine-tuning. CodeEditor correctly edits 1,113 programs while SOTA baselines can not work. The results prove that the superiority of our pre-training task and the pre-trained CodeEditor is more effective in automatic code editing., Comment: 18 pages, 6 figures

Cold goods manufacturers and logistics service providers are two essential groups of players in the goods safety issue in cold chains under the administration or inspection of governments and various stakeholders, including customers and final consumers. In this research, we applied the evolutionary game theory to examine the behavioral strategies of manufacturers and logistics service providers, while we formulated the governments and various other stakeholders’ impacts by contracted subsidy and penalty. First, we developed an evolutionary game theory model of the interaction between manufacturers and logistics service providers. Then, we examined the evolutionary stable strategy (ESS) of the manufacturers and logistics service providers under various constraints. Finally, we used simulation to demonstrate the impact of combinations of various parameters on the ESS and evolutionary paths. The results showed that the behavior strategies of the manufacturers and logistics service providers are interleaved and affected by the parameters in the developed model. We analyzed the ESSs and evolutionary paths by considering profits of the cold goods, the cold chain logistics costs, mainly the additional profits and costs of sharing information, and the subsidy and penalty regulated by contracts and governments. By tuning the parameters for numerical studies, we can find that the subsidy and penalty are essential for the cold chain manufactures and logistics service providers to adopt the information-sharing strategy, while the cost of the strategy and the profit of them constrains the positivity. Although, besides instant costs and profits, the information-sharing strategy can add values to cold chains in the long run, the administrators must consider the two populations of players and advocate them to adopt the information-sharing strategy consistently by using optimal policies.

In this paper, a weakly supervised framework is proposed for Abnormal Behavior Detection and Localization (ABDL) in the scenes. First, the objects in the scene such as pedestrians, vehicles, etc. are detected using the Faster Regional Conventional Neural Network (Faster R-CNN); then, the object behavior is described by a Histogram of Large Scale Optical Flow (HLSOF) descriptor; finally, the Multiple Instance Support Vector Machine (MISVM) is trained and then used to identify the testing behaviors as normal or abnormal. Summarily, the proposed approach has three main advantages: (1) Benefit from the Faster R-CNN, our approach can analyze the behavior at object-wise, which makes our approach has good generality and high computational efficiency; (2) The HLSOF descriptor can characterize the object behavior efficiently, and is insensitive to the variations of the size of objects; (3) As a weakly supervised learning framework, the MISVM only requires the labels at the bag level rather than instance level, which makes our approach has high accuracy as the supervised approaches but not requires completely labeled training samples, only the frame-level label is required. Experimental results analysis on different datasets validates the effectiveness of our approach.

As the great prevalence of various Internet of Things (IoT) terminals, how to solve the problem of isolated information among different IoT platforms attracts attention from both academia and industry. It is necessary to establish a trusted access system to achieve secure authentication and collaborative sharing. Therefore, this article proposes a distributed and trusted authentication system based on blockchain and edge computing, aiming to improve authentication efficiency. This system consists of physical network layer, blockchain edge layer and blockchain network layer. Through the blockchain network, an optimized practical Byzantine fault tolerance consensus algorithm is designed to construct a consortium blockchain for storing authentication data and logs. It guarantees trusted authentication and achieves activity traceability of terminals. Furthermore, edge computing is applied in blockchain edge nodes, to provide name resolution and edge authentication service based on smart contracts. Meanwhile, an asymmetric cryptography is designed, to prevent connection between nodes and terminals from being attacked. And a caching strategy based on edge computing is proposed to improve hit ratio. Our proposed authentication mechanism is evaluated with respect to communication and computation costs. Simulation results show that the caching strategy outperforms existing edge computing strategies by 6%–12% in terms of average delay, and 8%–14% in hit ratio.

Intracranial myeloid sarcoma is a very rare disease with poor prognosis. We report a case of a 28-year-old male patient who was admitted with intense headache, vision disturbance and severe vomiting in June 2017. He had a history of neurosurgical tumor resection operation in April 2017, and the pathological diagnosis was intracranial myeloid sarcoma. Bone marrow aspirate and biopsy had been conducted in May 2017, which demonstrated 5.5% blasts expressing CD13, CD33, CD34, CD117 and MPO, and the cytogenetic analysis demonstrated t(8;21)(q22;q22), and molecular studies showed a positive RUNX1-RUNX1T1 rearrangement. The diagnosis of acute myeloid leukemia (AML) with t (8; 21) (q22; q22)/RUNX1-RUNX1T1 was made, however, the patient refused to receive any systemic chemotherapy. Emergency cranial CT demonstrated a circular hyperdense mass (54mm×37mm), which was surrounded by hypodense peritumoral edema in the left cerebellar hemisphere, and the density of the lesions was uniform and the margin was clear. Idarubicin (12mg/m2·d×3 days) combined with high-dose cytarabine (2g/m2 q12h×3 days) was initiated for emergency chemotherapy. All of the above symptoms disappeared at the end of chemotherapy. On the first day after chemotherapy, the cranial CT indicated that the cranial lesion was markedly reduced (20mm×15mm), and on the sixth day after chemotherapy, the lesion was completely disappeared. Currently, there are no clear guidelines for the treatment of intracranial myeloid sarcoma, and our treatment approaches could provide a reference for this disease with such emergency situation.

Background Circular RNAs (circRNAs) have important functions in many fields of cancer biology. In particular, we previously reported that the oncogenic circRNA, circPRMT5, has a major role in bladder cancer progression. Therapy based on circRNAs have good prospects as anticancer strategies. While anti-circRNAs are emerging as therapeutics, the specific in vivo delivery of anti-circRNAs into cancer cells has not been reported and remains challenging. Methods Synthesized chrysotile nanotubes (SCNTs) with a relatively uniform length (~ 200 nm) have been designed to deliver an siRNA against the oncogenic circPRMT5 (si-circPRMT5) inhibit circPRMT5. In addition, the antitumor effects and safety evaluation of SCNTs/si-circPRMT5 was assessed with a series of in vitro and in vivo assays. Results The results showed that SCNTs/si-circPRMT5 nanomaterials prolong si-circPRMT5’s half-life in circulation, enhance its specific uptake by tumor cells, and maximize the silencing efficiency of circPRMT5. In vitro, SCNTs encapsulating si-circPRMT5 could inhibit bladder cancer cell growth and progression. In vivo, SCNTs/si-circPRMT5 inhibited growth and metastasis in three bladder tumor models (a subcutaneous model, a tail vein injection lung metastatic model, and an in situ model) without obvious toxicities. Mechanistic study showed that SCNTs/si-circPRMT5 regulated the miR-30c/SNAIL1/E-cadherin axis, inhibiting bladder cancer growth and progression. Conclusion The results highlight the potential therapeutic utility of SCNTs/si-circPRMT5 to deliver si-circPRMT5 to treat bladder cancer.

Oligonucleotide DNA aptamers represent an emergently important class of binding entities towards as different analytes as small molecules or even whole cells. Without the canonical isolation of individual aptamers following the SELEX process already the focused polyclonal libraries prepared by this in vitro evolution and selection can directly be used to label their dedicated analytes and to serve as binding molecules on surfaces. Here we report the first instance of a sensor able to discriminate between loaded and unloaded retinol binding protein 4 (RBP4), an important biomarker for the prediction of diabetes and kidney disease. The sensor relies purely on two aptamer libraries tuned such, that they discriminate between the protein isoforms, requiring no further sample labelling to detect RBP4 in both state. The evolution, binding properties of the libraries and the functionalization of graphene FET sensor chips are presented as well as the functionality of the resulting biosensor.

In more than 30 years of aptamer research, it has become widely accepted that aptamers are fascinating binding molecules for a vast variety of applications. However, the majority of targets have been proteins, although special variants of the so-called SELEX process for the molecular evolution of specific aptamers have also been developed, allowing for the targeting of small molecules as well as larger structures such as cells and even cellular networks of human (tumor) tissues. Although the provocative thesis is widely accepted in the field, that is, in principle, any level of complexity for SELEX targets is possible, the number of studies on whole organs or at least parts of them is limited. To pioneer this thesis, and based on our FluCell-SELEX process, here, we have developed polyclonal aptamer libraries against apices and the elongation/differentiation zones of plant roots as examples of organs. We show that dedicated libraries can specifically label the respective parts of the root, allowing us to distinguish them in fluorescence microscopy. We consider this achievement to be an initial but important evidence for the robustness of this SELEX variant. These libraries may be valuable tools for plant research and a promising starting point for the isolation of more specific individual aptamers directed against root-specific epitopes.

Based on their unique properties, oligonucleotide aptamers have been named a gift of biological chemistry to life science. We report the development of DNA aptamers as the first high-affinity binding molecules available for fast and rapid labeling of the human gut bacterium Akkermansia muciniphila with a certain impact on Alzheimer´s disease. Fast and reliable analyses of the composition of microbiomes is an emerging field in microbiology. We describe the molecular evolution and biochemical characterization of a specific aptamer library by a FluCell-SELEX and the characterization of specific molecules from the library by bioinformatics. The aptamer AKK13.1 exerted universal applicability in different analysis techniques in modern microbiology, including fluorimetry, confocal laser scanning microscopy and flow cytometry. It was also functional as a specific binding entity hybridized to anchor primers chemically coupled via acrydite-modification to the surface of a polyacrylamide-hydrogel, which can be prototypically used for the construction of affinity surfaces in sensor chips. Together, the performance and methodological flexibility of the aptamers presented here may open new routes not only to develop novel Akkermansia-specific assays for clinical microbiology and the analyses of human stool samples but may also be an excellent starting point for the construction of novel electronic biosensors., publishedVersion

Policy gradient methods are appealing in deep reinforcement learning but suffer from high variance of gradient estimate. To reduce the variance, the state value function is applied commonly. However, the effect of the state value function becomes limited in stochastic dynamic environments, where the unexpected state dynamics and rewards will increase the variance. In this paper, we propose to replace the state value function with a novel hindsight value function, which leverages the information from the future to reduce the variance of the gradient estimate for stochastic dynamic environments. Particularly, to obtain an ideally unbiased gradient estimate, we propose an information-theoretic approach, which optimizes the embeddings of the future to be independent of previous actions. In our experiments, we apply the proposed hindsight value function in stochastic dynamic environments, including discrete-action environments and continuous-action environments. Compared with the standard state value function, the proposed hindsight value function consistently reduces the variance, stabilizes the training, and improves the eventual policy., Accepted by IJCAI2021

Recent studies have demonstrated that changes in the abundance of the intestinal bacterium Blautia producta, a potential probiotic, are closely associated with the development of various diseases such as obesity, diabetes, some neurodegenerative diseases, and certain cancers. However, there is still a lack of an effective method to detect the abundance of B. producta in the gut rapidly. Especially, DNA aptamers are now widely used as biometric components for medical testing due to their unique characteristics, including high chemical stability, low production cost, ease of chemical modification, low immunogenicity, and fast reproducibility. We successfully obtained a high-affinity nucleic acid aptamer library (B.p-R14) after 14 SELEX rounds, which efficiently discriminates B. producta in different analysis techniques including fluorometric suspension assays or fluorescence microscopy from other major gut bacteria in complex mixtures and even in human stool samples. These preliminary findings will be the basis towards aptamer-based biosensing applications for the fast and reliable monitoring of B. producta in the human gut microbiome.

Archives information resources are the first-hand data and original records for recording and responding to emergencies. They play an important role in the emergency decision-making of emergencies. Strengthening the development of emergency archives information resources is an important way to improve emergency decision-making ability. The article takes the development of archives information resources for earthquake disasters in Sichuan Province as an example, conducts research on the development of archives information resources, and conducts case analysis to point out the current problems and optimization ideas.

Antarctic icefish Pagothenia borchgrevinki is an ideal model for studying heat stress mechanisms. The complete mitochondrial genome of P. borchgrevinki was sequenced in this study. The genome sequence is 17,299 bp in length, which comprises 13 protein-coding genes, 22 tRNAs, 2 rRNAs and a control region. The overall base composition is 20.45% G, 25.11% A, 29.46% T and 24.98 C%, with an A:T content of 54.57%.