Reinforcement Learning Based Quantization Strategy Optimal Assignment Algorithm for Mixed Precision

The quantization algorithm compresses the original network by reducing the numerical bit width of the model,which improves the computation speed. Because different layers have different redundancy and sensitivity to databit width. Reducing the data bit width will result in a loss of accuracy. Therefore, it is difficult to determinethe optimal bit width for different parts of the network with guaranteed accuracy. Mixed precision quantizationcan effectively reduce the amount of computation while keeping the model accuracy basically unchanged. In thispaper, a hardware-aware mixed precision quantization strategy optimal assignment algorithm adapted to low bitwidth is proposed, and reinforcement learning is used to automatically predict the mixed precision that meets theconstraints of hardware resources. In the state-space design, the standard deviation of weights is used to measurethe distribution difference of data, the execution speed feedback of simulated neural network accelerator inferenceis used as the environment to limit the action space of the agent, and the accuracy of the quantization model afterretraining is used as the reward function to guide the agent to carry out deep reinforcement learning training. Theexperimental results show that the proposed method obtains a suitable model layer-by-layer quantization strategyunder the condition that the computational resources are satisfied, and themodel accuracy is effectively improved.The proposed method has strong intelligence and certain universality and has strong application potential in thefield of mixed precision quantization and embedded neural network model deployment.

Machine Learning‑Enhanced Flexible Mechanical Sensing

To realize a hyperconnected smart society with high productivity,advances in flexible sensing technology are highly needed.Nowadays,flexible sensing technology has witnessed improvements in both the hardware performances of sensor devices and the data processing capabilities of the device’s software.Significant research efforts have been devoted to improving materials,sensing mechanism,and configurations of flexible sensing systems in a quest to fulfill the requirements of future technology.Meanwhile,advanced data analysis methods are being developed to extract useful information from increasingly complicated data collected by a single sensor or network of sensors.Machine learning(ML)as an important branch of artificial intelligence can efficiently handle such complex data,which can be multi-dimensional and multi-faceted,thus providing a powerful tool for easy interpretation of sensing data.In this review,the fundamental working mechanisms and common types of flexible mechanical sensors are firstly presented.Then how ML-assisted data interpretation improves the applications of flexible mechanical sensors and other closely-related sensors in various areas is elaborated,which includes health monitoring,human-machine interfaces,object/surface recognition,pressure prediction,and human posture/motion identification.Finally,the advantages,challenges,and future perspectives associated with the fusion of flexible mechanical sensing technology and ML algorithms are discussed.These will give significant insights to enable the advancement of next-generation artificial flexible mechanical sensing.

基于案例分析与翻转课堂的分析化学实验混合式教学实践

设计并实践了新的分析化学实验教学方法和考评方法。将案例分析与翻转课堂相结合,进行线上线下混合式教学。课前教师选择案例并设计问题,学生分组预习;课堂上学生讨论分析为主,教师查漏补缺为辅;课后学生撰写报告上传至线上平台,教师批阅并反馈问题。新的考评方法中,突出对学生表达能力、逻辑思维能力、团队协作能力、实验操作水平、实验台与实验室整洁水平的考查评价。实践表明,新的教学方法有助于提升学生的表达能力和实验操作水平,激发学生学习的积极性和挑战高难度学习任务的热情。新的考评方法为评价学生能力提供了可执行的标准。

Highly Sensitive Pseudocapacitive Iontronic Pressure Sensor with Broad Sensing Range

Flexible pressure sensors are unprecedentedly studied on monitoring human physical activities and robotics.Simultaneously,improving the response sensitivity and sensing range of flexible pressure sensors is a great challenge,which hinders the devices’practical application.Targeting this obstacle,we developed a Ti_(3)C_(2)T_(x)-derived iontronic pressure sensor(TIPS)by taking the advantages of the high intercalation pseudocapacitance under high pressure and rationally designed structural configuration.TIPS achieved an ultrahigh sen-sitivity(S_(min)>200 kPa^(−1),S_(max)>45,000 kPa^(−1))in a broad sensing range of over 1.4 MPa and low limit of detection of 20 Pa as well as stable long-term working durability for 10,000 cycles.The practical application of TIPS in physical activity monitoring and flexible robot manifested its versatile potential.This study provides a demonstration for exploring pseudocapacitive materials for building flexible iontronic sensors with ultrahigh sensitivity and sensing range to advance the development of high-performance wearable electronics.

Flexible Waterproof Piezoresistive Pressure Sensors with Wide Linear Working Range Based on Conductive Fabrics

Developing flexible sensors with high working performance holds intense interest for diverse applications in leveraging the Internet-of-things(IoT)infrastructures.For flexible piezoresistive sensors,traditionally most efforts are focused on tailoring the sensing materials to enhance the contact resistance variation for improving the sensitivity and working range,and it,however,remains challenging to simultaneously achieve flexible sensor with a linear working range over a high-pressure region(>100 kPa)and keep a reliable sensitivity.Herein,we devised a laserengraved silver-coated fabric as"soft"sensor electrode material to markedly advance the flexible sensor's linear working range to a level of 800 kPa with a high sensitivity of 6.4 kPa^-1 yet a fast response time of only 4 ms as well as long-time durability,which was rarely reported before.The integrated sensor successfully routed the wireless signal of pulse rate to the portable smartphone,further demonstrating its potential as a reliable electronic.Along with the rationally building the electrode instead of merely focusing on sensing materials capable of significantly improving the sensor's performance,we expect that this design concept and sensor system could potentially pave the way for developing more advanced wearable electronics in the future.

ASYMPTOTIC PROPERTIES OF A BRANCHING RANDOM WALK WITH A RANDOM ENVIRONMENT IN TIME

We consider a branching random walk in an independent and identically distributed random environment ξ=(ξn) indexed by the time. Let W be the limit of the martingale Wn=∫e^-txZn(dx)/Eξ∫e^-txZn(dx), with Zn denoting the counting measure of particles of generation n, and Eξ the conditional expectation given the environment ξ. We find necessary and sufficient conditions for the existence of quenched moments and weighted moments of W, when W is non-degenerate.

Discussion on Present Situation and Countermeasures about the development of nursing human resources

Nursing Human Resource Management is part of the human resources is nursing human resources by organizing nurses study, planning, training, and ultimately serve the medical career management stated objectives. Nursing human resource development of our human resources there are insufficient care, nurses configuration structure is irrational, the management system is not perfect, nurses are not high quality is the current status of nursing human resource mauagemenL and problems. This paper proposes a ＂human resource is the first resource＂ human resources management concept, has taken to develop a reasonable management mechanism, improve relevant laws and regulations of nursing human resources development, increase investment in education, strengthen education and training for nurses, the rational allocation of nursing human resources and other measures. In this paper, the connotation of Human Resources Management in the current situation as the starting point of care, from the current situation of nursing human resource development and the problems of the current situation of nursing human resource management and problems of nursing human resources management to discuss and propose appropriate for the development of human resource management solutions.

粉煤灰基土壤调理剂作用下盐碱土壤微观结构变化规律

以宁夏银北地区典型盐碱土壤为研究对象,采用X射线衍射(XRD)、红外光谱(FTIR)、比表面全自动物理吸附仪(BET)以及扫描电子显微镜(SEM)等测试技术对粉煤灰基土壤调理剂作用下盐碱土壤颗粒微观结构进行研究和系统表征。结果表明,利用调理剂改良后土壤矿物的主要成分未发生改变,但是添加调理剂后土壤有新晶相方解石生成;土壤的pH值大幅降低,由9.01降低到7.66;土壤红外吸收变化较小,Si-O-Si和有机质吸收峰略有增强;随着调理剂的增加,土壤小颗粒开始团聚,孔隙率逐渐增加,以添加2.5wt%组为例,15天后样品孔隙率达到21.30%,30天后样品孔隙率达到25.29%,约为未添加调理剂组的两倍;土壤颗粒的比表面积略有增加,30天后由初始的23.06 m;/g增加到25.55 m;/g。

Magnetically induced micropillar arrays for an ultrasensitive flexible sensor with a wireless recharging system

Significant efforts have been devoted to enhancing the sensitivity and working range of flexible pressure sensors to improve the precise measurement of subtle variations in pressure over a wide detection spectrum. However,achieving sensitivities exceeding 1000 kPa^(-1) while maintaining a pressure working range over 100 kPa is still challenging because of the limited intrinsic properties of soft matrix materials. Here, we report a magnetic field-induced porous elastomer with micropillar arrays(MPAs) as sensing materials and a well-patterned nickel fabric as an electrode. The developed sensor exhibits an ultrahigh sensitivity of 10,268 kPa^(-1)(0.6–170 kPa) with a minimum detection pressure of 0.25 Pa and a fast response time of 3 ms because of the unique structure of the MPAs and the textured morphology of the electrode. The porous elastomer provides an extended working range of up to 500 kPa with long-time durability. The sophisticated sensor system coupled with an integrated wireless recharging system comprising a flexible supercapacitor and inductive coils for transmission achieves excellent performance. Thus, a diverse range of practical applications requiring a low-to-high pressure range sensing can be developed. Our strategy, which combines a microstructured high-performance sensor device with a wireless recharging system, provides a basis for creating next-generation flexible electronics.

Stretchable and anti-impact iontronic pressure sensor with an ultrabroad linear range for biophysical monitoring and deep learning-aided knee rehabilitation

Monitoring biophysical signals such as body or organ movements and other physical phenomena is necessary for patient rehabilitation.However,stretchable flexible pressure sensors with high sensitivity and a broad range that can meet these requirements are still lacking.Herein,we successfully monitored various vital biophysical features and implemented in-sensor dynamic deep learning for knee rehabilitation using an ultrabroad linear range and highsensitivity stretchable iontronic pressure sensor(SIPS).We optimized the topological structure and material composition of the electrode to build a fully stretching on-skin sensor.The high sensitivity(12.43 kPa^(−1)),ultrabroad linear sensing range(1 MPa),high pressure resolution(6.4 Pa),long-term durability(no decay after 12000 cycles),and excellent stretchability(up to 20%)allow the sensor to maintain operating stability,even in emergency cases with a high sudden impact force(near 1 MPa)applied to the sensor.As a practical demonstration,the SIPS can positively track biophysical signals such as pulse waves,muscle movements,and plantar pressure.Importantly,with the help of a neuro-inspired fully convolutional network algorithm,the SIPS can accurately predict knee joint postures for better rehabilitation after orthopedic surgery.Our SIPS has potential as a promising candidate for wearable electronics and artificial intelligent medical engineering owing to its unique high signal-to-noise ratio and ultrabroad linear range.

Nickel-Catalyzed Regiodivergent Asymmetric Cycloadditions of α,β-Unsaturated Carbonyl Compounds

Regiodivergent asymmetric cycloadditions from the same set of starting materials offer interesting opportunities for rapid construction of optically active cyclic molecules with structural diversity.However,this remains a challenging task due to the difficulty of simultaneously controlling the regio-,diastereo-,and enantioselectivity in the ring formation processes.To address this long-standing problem.

Adaptively temporal graph convolution model for epidemic prediction of multiple age groups

Introduction:Multivariate time series prediction of infectious diseases is significant to public health,and the deep learning method has attracted increasing attention in this research field.Material and methods:An adaptively temporal graph convolution(ATGCN)model,which leams the contact patterns of multiple age groups in a graph-based approach,was proposed for COVID-19 and influenza prediction.We compared ATGCN with autoregressive models,deep sequence learning models,and experience-based ATGCN models in short-term and long-term prediction tasks.Results:Results showed that the ATGCN model performed better than the autoregressive models and the deep sequence learning models on two datasets in both short-term(12.5%and 10%improvements on RMSE)and longterm(12.4%and 5%improvements on RMSE)prediction tasks.And the RMSE of ATGCN predictions fluctuated least in different age groups of COVID-19(0.029±0.003)and influenza(0.059±0.008).Compared with the Ones-ATGCN model or the Pre-ATGCN model,the ATGCN model was more robust in performance,with RMSE of 0.0293 and 0.06 on two datasets when horizon is one.Discussion:Our research indicates a broad application prospect of deep learning in the field of infectious disease prediction.Transmission characteristics and domain knowledge of infectious diseases should be further applied to the design of deep learning models and feature selection.Conclusion:The ATGCN model addressed the multivariate time series forecasting in a graph-based deep learning approach and achieved robust prediction on the confirmed cases of multiple age groups,indicating its great potentials for exploring the implicit interactions of multivariate variables.