Optimized operation scheme of flash-memory-based neural network online training with ultra-high endurance

With the rapid development of machine learning,the demand for high-efficient computing becomes more and more urgent.To break the bottleneck of the traditional Von Neumann architecture,computing-in-memory(CIM)has attracted increasing attention in recent years.In this work,to provide a feasible CIM solution for the large-scale neural networks(NN)requiring continuous weight updating in online training,a flash-based computing-in-memory with high endurance(10^(9) cycles)and ultrafast programming speed is investigated.On the one hand,the proposed programming scheme of channel hot electron injection(CHEI)and hot hole injection(HHI)demonstrate high linearity,symmetric potentiation,and a depression process,which help to improve the training speed and accuracy.On the other hand,the low-damage programming scheme and memory window(MW)optimizations can suppress cell degradation effectively with improved computing accuracy.Even after 109 cycles,the leakage current(I_(off))of cells remains sub-10pA,ensuring the large-scale computing ability of memory.Further characterizations are done on read disturb to demonstrate its robust reliabilities.By processing CIFAR-10 tasks,it is evident that~90%accuracy can be achieved after 109 cycles in both ResNet50 and VGG16 NN.Our results suggest that flash-based CIM has great potential to overcome the limitations of traditional Von Neumann architectures and enable high-performance NN online training,which pave the way for further development of artificial intelligence(AI)accelerators.

Interaction Mechanism between Formation Process of Bank Collapse Disaster Chain and Territorial Space Utilization of Large Reservoirs

Number of reservoirs in China ranks the first in the world. Due to the complex geology, and superimposing rainfall and reservoir water fluctuation, the bank collapse chain is prone to disasters. The Yangtze River Reservoir is key geological disaster prevention area. Studying the process of reservoir disaster is significant because of the limited territorial space utilization. Scientific and technological issues, i.e., the mechanism of bank collapse disaster chain of large reservoirs, the interaction mechanism of bank collapse disaster chain and territorial space utilization, the early identification, monitoring technology and ecological prevention and control technology system of disaster chain, and the territorial space geological safety and control technology system are focused. We consider the material transformation, energy transfer and information transmission in disaster chain;adopt the survey, Space-Air-Ground integrated monitoring, theoretical analysis, numerical simulation and the multidisciplinary research methods;reveal the chain source development, evolution process of secondary and derivative disasters;explore the interaction mechanism of disaster chain and territorial space utilization;construct the system of early identification, monitoring, early warning, control and ecological preven-tion to achieve Emission Peak and Carbon Neutrality;provide theoretical and technical support for the territorial space geological safety, regulation and utilization of large reservoirs.

哺乳动物卵母细胞不对称分裂的研究进展

哺乳动物卵母细胞成熟过程需要进行两次连续的不对称分裂,最终形成体积差异巨大的子细胞:大体积的卵母细胞和两种体积较小的极体。不对称分裂现象是哺乳动物卵母细胞减数分裂的典型特征,不对称分裂后的卵母细胞是高度极化的细胞。精卵结合后,细胞重新恢复了对称分裂,但是在卵母细胞减数分裂过程中形成的极性特征却得以保留并影响早期胚胎的极性。本文对近年来在哺乳动物卵母细胞不对称分裂方面的相关研究展开综述,从细胞质不对称分裂和细胞核不对称分裂两个方面对染色体、细胞骨架在哺乳动物卵母细胞不对称分裂中的作用、细胞器在哺乳动物卵母细胞成熟过程中的重组分配、染色体非随机分离等过程进行介绍,旨在从细胞和分子水平阐述哺乳动物卵母细胞不对称分裂的主要机制。

Ultrasmall AuPd nanoclusters on amine-functionalized carbon blacks as high-performance bi-functional catalysts for ethanol electrooxidation and formic acid dehydrogenation

The synthesis of ultrasmall metal nanoclusters(NCs) with high catalytic activities is of great importance for the development of clean and renewable energy technologies but remains a challenge. Here we report a facile wet-chemical method to prepare ~1.0 nm Au Pd NCs supported on amine-functionalized carbon blacks. The Au Pd NCs exhibit a specific activity of 5.98 mA cm_(AuPd)^(-2)and mass activity of 5.25 A mg_(auPd)^(-1) for ethanol electrooxidation, which are far better than those of commercial Pd/C catalysts(1.74 mAcm_(AuPd)^(-2) and 0.54 A mg_(Pd)^(-1) ). For formic acid dehydrogenation, the Au Pd NCs have an initial turn over frequency of 49339 h^(-1) at 298 K without any additive, which is much higher than those obtained for most of reported Au Pd catalysts. The reported synthesis may represent a facile and low-cost approach to prepare other ultrasmall metal NCs with high catalytic activities for various applications.

Flash-based in-memory computing for stochastic computing in image edge detection

The“memory wall”of traditional von Neumann computing systems severely restricts the efficiency of data-intensive task execution,while in-memory computing(IMC)architecture is a promising approach to breaking the bottleneck.Although variations and instability in ultra-scaled memory cells seriously degrade the calculation accuracy in IMC architectures,stochastic computing(SC)can compensate for these shortcomings due to its low sensitivity to cell disturbances.Furthermore,massive parallel computing can be processed to improve the speed and efficiency of the system.In this paper,by designing logic functions in NOR flash arrays,SC in IMC for the image edge detection is realized,demonstrating ultra-low computational complexity and power consumption(25.5 fJ/pixel at 2-bit sequence length).More impressively,the noise immunity is 6 times higher than that of the traditional binary method,showing good tolerances to cell variation and reliability degradation when implementing massive parallel computation in the array.

Understanding the growth mechanisms of metal-based core–shell nanostructures revealed by in situ liquid cell transmission electron microscopy

Metal-based core-shell nanostructures have garnered enduring interest due to their unique properties and functionalities.However,their growth and transformation mechanisms in liquid media remain largely unknown because they lack direct observation of the dynamic growth process with high spatial and temporal resolution.Developing the in situ liquid cell transmission electron microscopy(TEM)technique offers unprecedented real-time imaging and spectroscopy capabilities to directly track the evolution of structural and chemical transformation of metal-based core–shell nanostructures in liquid media under their working condition.Here,this review highlights recent progress in utilizing in situ liquid cell TEM characterization technique in investigating the dynamic evolution of material structure and morphology of metal-based core–shell nanostructures at the nano/atomic scale in real-time.A brief introduction of the development of liquid cells for in situ TEM is first given.Subsequently,recent advances in in situ liquid cell TEM for the fundamental study of growth mechanisms of metal based core–shell nanostructures are discussed.Finally,the challenge and future developments of metalbased core–shell nanostructures for in situ liquid cell TEM are proposed.Our review is anticipated to inspire ongoing interest in revealing unseen growth dynamics of core–shell nanostructures by in situ liquid cell TEM technique.

Wireless Communication Based on Microwave Photon-Level Detection with Superconducting Devices:Achievable Rate Prediction

Future wireless communication systemembraces physical-layer signal detection with highsensitivity, especially in the microwave photon level.Currently, the receiver primarily adopts the signal detection based on semi-conductor devices for signal detection, while this paper introduces high-sensitivityphoton-level microwave detection based on superconducting structure. We first overview existing works onthe photon-level communication in the optical spectrum as well as the microwave photon-level sensingbased on superconducting structure in both theoreticaland experimental perspectives, including microwavedetection circuit model based on Josephson junction,microwave photon counter based on Josephson junction, and two reconstruction approaches under background noise. In addition, we characterize channelmodeling based on two different microwave photondetection approaches, including the absorption barrierand the dual-path Handury Brown-Twiss (HBT) experiments, and predict the corresponding achievablerates. According to the performance prediction, it isseen that the microwave photon-level signal detectioncan increase the receiver sensitivity compared withthe state-of-the-art standardized communication system with waveform signal reception, with gain over 10dB.

Challenges of Employing VoiceThread to Enhance U.S.Collegiate Chinese Language Learners’Oral Proficiency

The digital age in education inspires Chinese teachers to integrate Web 2.0 technologies into Chinese as a second language(CSL)curricula to enhance teaching effects.However,few studies examine the challenges that teachers face when integrating Web 2.0 technologies to improve student oral proficiency in Chinese.This study aims to investigate students’perceptions of the effectiveness of the Web 2.0 technology VoiceThread(VT).It examines the challenges of integrating VT to enhance student oral proficiency.The authors conducted two studies with 31 students responding to a questionnaire in Study 1 and 48 students responding to the same questionnaire in Study 2.All students are beginner-level CSL learners.Three results were obtained.Result 1 shows that documents and articles report VT as popular and easy to use,however,few teachers employed VT in their teaching practices.Result 2 indicates another discrepancy.Although students highly evaluated the usefulness of the VT-facilitated activities,they were reluctant to perform these activities.By comparing Study 1 and Study 2,the authors obtained Result 3:Students in Study 2 evaluated VT-integrated teaching practices consistently higher than those in Study 1.Teachers’using experiences primarily led these perception differences.This study provides CSL teachers with a vision of the opportunities and challenges that may occur when they integrate Web 2.0 technologies into curricular to improve CSL oral proficiency.

Active molecule-based theranostic agents for tumor vasculature normalization and antitumor efficacy

Tumor vascular normalization has emerged as a promising strategy for synergistic therapy recently.Based on the strategy of“fluorescence turn on-controllable release”,a novel bifunctional candidate was con-structed based on previous developed vascular normalization inducer QDAU5,which could self-assemble to form functional enzyme infrared QDAU5 nanoparticles(FEIRQ NPs).Subsequently,biological evaluation demonstrated that the FEIRQ NPs could induce ferroptosis,endoplasmic reticulum stress,and antigen pre-conditioning and maturation of dendritic cells and CD8^(+)T cells,leading to excellent antitumor efficacy in the absence of cytotoxic drugs.Additionally,FEIRQ NPs show high fluorescence intensity upon expo-sure to theβ-galactosidase(β-Gal)enzyme expressed in ovarian cancer,enabling real-time monitoring of therapeutic effects.Overall,our findings suggest a prospering strategy to early diagnosis and efficient therapy for ovarian cancer without cytotoxicity.

A novel preimplantation genetic testing strategy for a subtelomeric genetic disorder:A case study

Preimplantation genetic testing(PGT)is a precise and effective technique for detecting inherited pathogenic mutations to prevent birth defects.However,there are few reports on PGT for pseudo-autosomal genetic diseases.

Synergistic Combination of Biodegradable Peptide Polymer and Curcumin as Promising Antibiotic Substitution in Aquaculture to Alleviate the Global Challenge of Antimicrobial Resistance

The massive use of antibiotics in aquaculture and resulted antibacterial resistance problem urgently need antibiotic substitutions. Herein, we report a promising substitution of aquaculture antibiotics using a synergistic combination of biodegradable peptide polymers and curcumin, a natural compound from plant. The synergistic combination shows strong antibacterial activity against V. fluvialis and some other common bacteria in aquaculture. The membrane-damaging antibacterial mechanism echoes our finding that the synergistic combination will not induce bacteria to develop resistance after continuous use. The synergistic combination also displays effective cure on V. fluvialis-infected zebrafish. The biodegradability of the peptide polymer enables the combination to lose antibacterial activity and will not cause selective pressure on bacterial in the environment. Our study indicates potential application of synergistic composition, biodegradable peptide polymer and curcumin, as promising antibiotic substitution in aquaculture, which represents a promising strategy to address the global challenge of antimicrobial resistance.

GPR160 is a potential biomarker associated with prostate cancer

Dear Editor,Prostate cancer(PC)is one of the most common noncutaneous cancers among men worldwide with a relatively higher incidence and mortality rate.1 Conventional screening for prostate cancer relies on prostate-specific antigen(PSA),a valuable biomarker but has some deficiencies.Radical prostatectomy,androgen ablation,and radiotherapy are still commonly used treatments for localized prostate cancer.

Tracking the atomic pathways of Pt3Ni-Ni(OH)2 core-shell structures at the gas-liquid interface by in-situ liquid cell TEM

Using the in-situ liquid cell transmission electron microscopy, the three-stage growth of Pt3Ni-Ni(OH)2 core-shell structures at the gas-liquid interfaces was clearly observed, which consists of(1) a thermodynamically driven Pt3Ni alloy core by the monomer attachment,(2) a nickel(Ni) shell formation due to the depletion of the Pt salt precursor, and(3) the oxidation and of the Ni shell into Ni(OH)2 flakes. We also further observed the nucleation and growth of the Ni(OH)2 flakes on an existing layer either at the middle part or at the step edge. More interestingly, the dynamic transformation among a Pt3Ni alloy, Ni clusters and Ni(OH)2 flakes was also imaged even at a high electron dose rate.

Deactivation of Cu-SAPO-34 by urea-related deposits at low temperatures and the regeneration

Selective catalytic reduction(SCR) with urea catalyzed by Cu-SAPO-34 is an effective method to eliminate NO_x from diesel exhaust. However, urea-related deposits may form during cold-start and urban driving due to low exhaust temperatures. The activity of CuSAPO-34 at 175°C is significantly degraded by urea exposure, and 300°C is required for regeneration. Through in-situ diffuse reflectance infrared Fourier transform spectroscopy(DRIFTS) and temperature-programmed hydrolysis studies, the dominant stable deposit at 175°C is identified as biuret, which can be eliminated at 300°C. The urea-derived deactivation and regeneration mechanisms of Cu-SAPO-34 were compared with those of anatase-supported catalysts.

In-situ liquid cell TEM investigation on assembly and symmetry transformation of Pt superlattice

Two dimensional(2D)nanocrystal functional superlattices with a well controlled structure are of significant importance in photonic,plasmonic and optoelectronic applications and have been well studied,but it remains challenging to understand the formation mechanism and development pathway of the superlattice.In this study,we employed in-situ liquid cell transmission electron microscopy to study the formation of 2D superlattice and its local phase transition from hexagonal-to-square nanocrystal ordering.When colloidal nanocrystals flowed in the solution,long-range ordered hexagonal superlattice could be formed either through shrinking and rearrangement of nanocrystal aggregates or via nanocrystal attachment.As the nanocrystals’shape transformed from truncated octahedral to cube,the local superlattice rearranged to square geometry.Moreover,our observations and quantitative analyses reveal that the phase transition from hexagonal to square mainly originates from the strong van der Waals interactions between the vertical(100)facets.The tracking of 2D cube superlattice formation in real-time could provide unique insights on the governing force of superlattice assembling and stabilization.

Integrated facial analysis and targeted sequencing identifies a novel KDM6A pathogenic variant resulting in Kabuki syndrome

Kabuki syndrome(KS)is a rare congenital mental retardation condition characterized by facial dysmorphia,visceral and skeletal malformations,and developmental delay.The integrated phenotype and genotype-based prioritization is critical for diagnoses of genetic diseases.In this study,a Chinese woman,presenting with characteristic facial features of KS,came for pre-pregnancy consultation.We aimed to clarify the diagnosis and provide pre-pregnancy genetic counseling.Facial dysmorphology analysis and next-generation sequencing-based multigene panel approach were used to identify candidate syndromes and causative variants,respectively.The candidate variant was verified by Sanger sequencing.We identified a novel de novo KDM6A pathogenic variant(c.3521G>A)in the woman,which was in line with the Face2Gene analysis result.Peripheral blood RNA assay showed that the variant transcript underwent the nonsense-mediated mRNA decay and led to subsequent haploinsufficiency of KDM6A.Our study provides the genetic diagnosis method for KS type 2 and identifies the first KDM6A point variant in Chinese patient.

Structural Origins for Enhanced Thermal Stability and Glass‑Forming Ability of Co–B Metallic Glasses with Y and Nb Addition

The effects of Y and Nb addition on thermal stability,glass-forming ability(GFA),and magnetic softness of Co75B25 metallic glass(MG)were comprehensively investigated.The experimental results indicated that the thermal stability,GFA,and magnetic softness of the studied MGs increase in the order Co_(75)B_(25)<Co_(73)Nb_(2)B_(25)<Co_(71.5)Y_(3.5)B_(25)<Co_(69.5)Y_(3.5)Nb_(2)B_(25).The structural origins of the improved properties were revealed by ab initio molecular dynamics(AIMD)simulations and density functional theory(DFT)calculations.Results showed that the B-centered prism units are the primary structure-forming units of the four MGs,connect through vertex-,edge-,and face-shared(VS,ES,and FS)atoms,and Co-centered units tend to connect with Co/B-centered units via the intercross-shared(IS)atoms.The addition of Y and Nb not only plays the role of connecting atoms but also enhances both bond strengths and the fractions of icosahedral-like units in increasing order Co_(75)B_(25)<Co_(73)Nb_(2)B_(25)<Co_(71.5)Y_(3.5)B_(25)<Co_(69.5)Y_(3.5)Nb_(2)B_(25),which is conducive to the enhancement of the structural stability,atomic packing density,and viscosity,thereby improving thermal stability and GFA.In addition,the improvement of structural stability and homogeneity leads to enhanced magnetic softness.

Atomic mechanisms of hexagonal close-packed Ni nanocrystallization revealed by in situ liquid cell transmission electron microscopy

The fundamental understanding of the mechanism underlying the early stages of crystallization of hexagonal-close-packed(hcp)nanocrystals is crucial for their synthesis with desired properties,but it remains a significant challenge.Here,we report using in situ liquid cell transmission electron microscopy(TEM)to directly capture the dynamic nucleation process and track the real-time growth pathway of hcp Ni nanocrystals at the atomic scale.It is demonstrated that the growth of amorphous-phase-mediated hcp Ni nanocrystals is from the metal-rich liquid phases.In addition,the reshaped preatomic facet development of a single nanocrystal is also imaged.Theoretical calculations further identify the non-classical features of hcp Ni crystallization.These discoveries could enrich the nucleation and growth model theory and provide useful information for the rational design of synthesis pathways of hcp nanocrystals.

An average-value-at-risk criterion for Markov decision processes with unbounded costs

We study the Markov decision processes under the average-value-at-risk criterion.The state space and the action space are Borel spaces,the costs are admitted to be unbounded from above,and the discount factors are state-action dependent.Under suitable conditions,we establish the existence of optimal deterministic stationary policies.Furthermore,we apply our main results to a cash-balance model.

Efficient Modulation of Exon Skipping via Antisense Circular RNAs

Splice-switching antisense oligonucleotides(ASOs)and engineered U7 small nuclear ribonucleoprotein(U7 Sm OPT)are the most commonly used methods for exon skipping.However,challenges remain,such as limited organ delivery and repeated dosing for ASOs and unknown risks of by-products produced by U7 Sm OPT.Here,we showed that antisense circular RNAs(AS-circRNAs)can effectively mediate exon skipping in both minigene and endogenous transcripts.We also showed a relatively higher exon skipping efficiency at the tested Dmd minigene than U7 Sm OPT.AS-circRNA specifically targets the precursor mRNA splicing without off-target effects.Moreover,AS-circRNAs with adeno-associated virus(AAV)delivery corrected the open reading frame and restored the dystrophin expression in a mouse model of Duchenne muscular dystrophy.In conclusion,we develop an alternative method for regulating RNA splicing,which might be served as a novel tool for genetic disease treatment.