Processing Mechanisms Research Articles

Robust iterative value conversion: Deep reinforcement learning for neurochip-driven edge robots

A neurochip is a device that reproduces the signal processing mechanisms of brain neurons and calculates Spiking Neural Networks (SNNs) with low power consumption and at high speed. Thus, neurochips are attracting attention from edge robot applications, which suffer from limited battery capacity. This paper aims to achieve deep reinforcement learning (DRL) that acquires SNN policies suitable for neurochip implementation. Since DRL requires a complex function approximation, we focus on conversion techniques from Floating Point NN (FPNN) because it is one of the most feasible SNN techniques. However, DRL requires conversions to SNNs for every policy update to collect the learning samples for a DRL-learning cycle, which updates the FPNN policy and collects the SNN policy samples. Accumulative conversion errors can significantly degrade the performance of the SNN policies. We propose Robust Iterative Value Conversion (RIVC) as a DRL that incorporates conversion error reduction and robustness to conversion errors. To reduce them, FPNN is optimized with the same number of quantization bits as an SNN. The FPNN output is not significantly changed by quantization. To robustify the conversion error, an FPNN policy that is applied with quantization is updated to increase the gap between the probability of selecting the optimal action and other actions. This step prevents unexpected replacements of the policy’s optimal actions. We verified RIVC’s effectiveness on a neurochip-driven robot. The results showed that RIVC consumed 1/15 times less power and increased the calculation speed by five times more than an edge CPU (quad-core ARM Cortex-A72). The previous framework with no countermeasures against conversion errors failed to train the policies. Videos from our experiments are available: https://youtu.be/Q5Z0-BvK1Tc.

Effect of process parameters on the mechanical properties of wires produced from A356 aluminum alloy chips by Continuous Friction Stir Extrusion: Experiments and numerical simulation

Recycling of metals is becoming crucial from an economic and environmental point of view. The solid-state recycling process Continuous Friction Stir Extrusion was used to produce wires out of A356-T6 chips. The mechanical properties of the produced wires were explored by varying the main process parameters. Characterization involved Vickers hardness tests, tensile tests, grain size measurements, and fracture surface analysis. It has been found that it is possible to achieve 77 % of the Ultimate Tensile Strength (UTS) and 92 % of Vickers hardness with respect to the as-fabricated A356 alloy. The average grain size increases with the tool rotational with values ranging from about 9 µm to about 11 µm. A 3D dedicated numerical model was used to predict the distributions and histories of primary field variables, and to calculate the Piwnik-Plata parameter, fostering a more in-depth understanding of the process mechanics. This allows for the precise prediction of unacceptable product quality of the bonding when the Plata and Piwnik parameters are low. Predicted temperature close to the rotating tool should reach 400 °C while the cochlea temperature should be below 100 °C for sound wires production thus avoiding early chip bonding and process failure.

Towards a better understanding of the surface smoothing effect in gas cluster ion beam processing with molecular dynamics simulation and experiment.

In recent years, the application of gas cluster ion beam (GCIB) technology has made great progress. Due to the similar essence of a monoatomic ion beam, the GCIB also shows flashes of brilliance in material processing. It has been reported that smoothness can be greatly improved after the rough surface is bombarded by the GCIB. This indicates that the GCIB processing has great potential in optical fabrication. Although the surface smoothing effect has been investigated, there is still a lack of dynamic micro-analysis for GCIB processing, which is limited for better understanding the mechanism of smoothing effect. In this paper, the surface smoothing effect in GCIB processing is explicitly investigated with molecular dynamics (MD) simulation and experiment. The principle of GCIB processing is compared with the traditional monoatomic ion-beam based processing, and details of the MD simulation procedure are introduced. Based on this, the dynamic micro-analysis of GCIB processing is conducted under different processing conditions. The simulations reveal the phenomena of atomic removal and migration in GCIB processing, which plays an important role in explaining the mechanism of surface smoothing effect. The experiment was performed on the silicon substrate with the in-house GCIB processing machine. The results indicate that the initial rough surface with dense protrusions can be greatly smoothed, and the root mean square (RMS) value is reduced from 0.586 nm to 0.191 nm. Both simulation and experiment can provide a better understanding of smoothing effect mechanism in GCIB processing.

An Insight into the Mechanism of DNA Cleavage by DNA Endonuclease from the Hyperthermophilic Archaeon Pyrococcus furiosus.

Hyperthermophilic archaea such as Pyrococcus furiosus survive under very aggressive environmental conditions by occupying niches inaccessible to representatives of other domains of life. The ability to survive such severe living conditions must be ensured by extraordinarily efficient mechanisms of DNA processing, including repair. Therefore, in this study, we compared kinetics of conformational changes of DNA Endonuclease Q from P. furiosus during its interaction with various DNA substrates containing an analog of an apurinic/apyrimidinic site (F-site), hypoxanthine, uracil, 5,6-dihydrouracil, the α-anomer of adenosine, or 1,N6-ethenoadenosine. Our examination of DNA cleavage activity and fluorescence time courses characterizing conformational changes of the dye-labeled DNA substrates during the interaction with EndoQ revealed that the enzyme induces multiple conformational changes of DNA in the course of binding. Moreover, the obtained data suggested that the formation of the enzyme-substrate complex can proceed through dissimilar kinetic pathways, resulting in different types of DNA conformational changes, which probably allow the enzyme to perform its biological function at an extreme temperature.

A feature detection network based on self-attention mechanism for underwater image processing

Underwater image processing is seriously challenged by the absorption and scattering of light in water. This paper introduces a self supervised learning network that utilizes self attention mechanism for underwater visual applications. The goal is to improve the effectiveness and stability of underwater image feature detection. The combination of self attention mechanism enhances the sensitivity of the original network architecture to degraded features in blurred underwater images. A modified underwater image synthesis method is proposed for network training and evaluation to address the difficulties and high costs of obtaining underwater images. And a step-by-step training method was proposed, proving the effectiveness of this training method. The proposed network is trained using a step-by-step method and evaluated on various underwater image datasets. The experimental results show that the algorithm outperforms other methods in distribution, quantitative, and qualitative measurements, and exhibits enhanced feature detection ability in blurred images as water turbidity increases.

Attentional dynamics of evidence accumulation explain why more numerate people make better decisions under risk

In decisions under risk, more numerate people are typically more likely to choose the option with the highest expected value (EV) than less numerate ones. Prior research indicates that this finding cannot be explained by differences in the reliance on explicit EV calculation. The current work uses the attentional Drift Diffusion Model as a unified computational framework to formalize three candidate mechanisms of pre-decisional information search and processing—namely, attention allocation, amount of deliberation, and distorted processing of value—which may differ between more and less numerate people and explain differences in decision quality. Computational modeling of an eye-tracking experiment on risky choice demonstrates that numeracy is linked to how people allocate their attention across the options, how much evidence they require before committing to a choice, and also how strongly they distort currently non-attended information during preference formation. Together, especially the latter two mechanisms largely mediate the effect of numeracy on decision quality. Overall, the current work disentangles and quantifies latent aspects of the dynamics of preference formation, explicates how their interplay may give rise to manifest differences in decision quality, and thereby provides a fully formalized, mechanistic explanation for the link between numeracy and decision quality in risky choice.

Study on the polishing performance and mechanism of sapphire wafers by different types of degradable surfactants

Traditional chemical-mechanical polishing (CMP) slurry suffers from significant environmental pollution, and easy reaggregation of ultra-fine abrasives, all of which hinder its further application. In this study, the processing performance and mechanism of three different degradable surfactants to sapphire wafers were investigated, utilizing degradable aminomethylpropanol and xylitol as pH regulator and complexing agent of polishing slurry, respectively, which is aimed at solving the problem of the easy agglomeration of ultra-fine abrasives while achieving the super-precision and pollution-free process of sapphire wafers. After polishing, the surface roughness (Sa) of sapphire wafers decreases from 7.240 nm to 0.396 nm, which is 44.8 % lower than that without surfactant. Only <1 nm is the PV value, and there is a 12.9 % increase in the material removal rate. The action mechanism of biodegradable surfactants during CMP was revealed through TEM, XPS, electrochemical testing, contact angle measurement, and particle size analysis. The results showed that surfactants can effectively improve the wettability of the polishing slurry on the wafer and the dispersity of the ultra-fine abrasives in the slurry, thus facilitating the interfacial reaction between the sapphire surface and polishing slurry and the homogeneity of the material removal. This work truly realizes a green and pollution-free process that meets the requirements of ultra-smooth and efficient polishing of sapphire wafers.

The More, the Better? Exploring the Effects of Modal and Codal Redundancy on Learning and Cognitive Load: An Experimental Study

This study explores how receiving identical information from different sources affects learning and cognitive load, focusing on two types of redundancy: modal redundancy, where redundant information comes from two visual sources (images and written text), and codal redundancy, where redundant information comes from two sources of different modalities which utilize the same symbol system (spoken and written text). Using a 2 × 2 between-subjects design involving modal (yes/no) and codal (yes/no) redundancy, 158 participants completed twenty learning tasks, consisting of ten construction and ten recall tasks. Additionally, they rated their cognitive load by indicating their perceived task difficulty and mental load. Overall, results indicate positive main effects of modal redundancy and negative effects of codal redundancy on learning and cognitive load. Furthermore, significant interaction effects suggest that modal redundancy may counterbalance the negative effects of codal redundancy, implying a compensatory mechanism in cognitive processing for construction tasks. These results highlight the importance of considering both modal and codal redundancy and their interaction in instructional design.

Processing of hydrocarbon dust in star-forming galaxies revealed with AKARI

Abstract Hydrocarbon dust is one of the dominant components of interstellar dust, which mainly consists of polycyclic aromatic hydrocarbons and aliphatic hydrocarbons. While hydrocarbon dust is thought to be processed in interstellar radiation fields or shocks, detailed processing mechanisms are not completely understood yet. We investigate the processing of hydrocarbon dust by analyzing the relation between the luminosities emitted by hydrocarbon dust and the total infrared luminosities $(L_{\mathrm{IR}})$ for 138 star-forming galaxies at redshift $z \lt 0.3$. Using near-infrared 2.5–5$\, \mu {\rm m}$ spectra obtained with AKARI, we derived the luminosities of the aromatic hydrocarbon feature at 3.3$\, \mu {\rm m}$ ($L_\mathrm{aromatic}$) and the aliphatic hydrocarbon feature at 3.4–3.6$\, \mu {\rm m}$ ($L_\mathrm{aliphatic}$). We also derived $L_\mathrm{IR}$ and the radiation field strength by modeling the spectral energy distributions of the 138 galaxies with AKARI, WISE, and IRAS photometry data. We find that galaxies with higher $L_\mathrm{IR}$ tend to exhibit lower $L_\mathrm{aliphatic}/L_\mathrm{aromatic}$ ratios. Furthermore, we find that there is an anti-correlation between $L_\mathrm{aliphatic}/L_\mathrm{aromatic}$ ratios and the radiation field strength, and also that the galaxies with low $L_\mathrm{aliphatic}/L_\mathrm{aromatic}$ ratios are dominated by merger galaxies. These results support the suggestion that hydrocarbon dust is processed through photodissociation in strong radiation fields and/or shocks during merging processes of galaxies; the $L_\mathrm{aliphatic}/L_\mathrm{aromatic}$ ratio is likely to decrease in such harsh interstellar conditions since the aliphatic bonds are known to be chemically weaker than the aromatic bonds.

Human pose estimation in complex background videos via Transformer-based multi-scale feature integration

Human posture estimation is still a hot research topic. Previous algorithms based on traditional machine learning have difficulties in feature extraction and low fusion efficiency. To address these problems, we proposed a Transformer-based method. We combined three techniques, namely the Transformer-based feature extraction module, the multi-scale feature fusion module, and the occlusion processing mechanism, to capture the human pose. The Transformer-based feature extraction module uses the self-attention mechanism to extract key features from the input sequence, the multi-scale feature fusion module fuses feature information of different scales to enhance the perception ability of the model, and the occlusion processing mechanism can effectively handle occlusion in the data and effectively remove background interference. Our method has shown excellent performance through verification on the standard dataset Human3.6M and the wild video dataset, achieving accurate pose prediction in both complex actions and challenging samples.

In-line monitoring and modelling of particle size reduction of hexagonal boron nitride (hBN) suspension in femtosecond laser processing

This study investigates the particle-size reduction of hexagonal boron nitride (hBN) suspensions under femtosecond-laser irradiation. We propose an in-line monitoring method that uses transmission spectroscopy (TS) in the visible wavelength region to evaluate the changes in particle-size distribution. The Ångström exponent (α), derived from the wavelength dependence of optical thickness, is employed as an indicator of particle-size reduction. The effectiveness of α is validated through comparison with particle-size distributions obtained by the particle tracking method. We demonstrate real-time monitoring of particle-size variation in a flow system using in-line TS measurements. The dependence of α on laser energy and hBN concentration is also investigated. Furthermore, we introduce an indicator, the average number of size-reduction events per particle (N SR), and a model for estimating it under various experimental conditions. The N SR indicator and the model for its derivation provide a unified understanding of the experimental results obtained under various conditions, such as treatment volume, laser energy, and hBN concentration, suggesting that they capture the essential aspects of laser propagation in the suspension and the particle-size reduction process. This study highlights the potential of α as an in-line process monitoring tool and the significance of the N SR indicator and its model in understanding the underlying mechanisms of laser processing for particle-size reduction in suspensions.

Coordination between midcingulate cortex and retrosplenial cortex in pain regulation.

The cingulate cortex, with its subregions ACC, MCC, and RSC, is key in pain processing. However, the detailed interactions among these regions in modulating pain sensation have remained unclear. In this study, chemogenetic tools were employed to selectively activate or inhibit neuronal activity in the MCC and RSC of rodents to elucidate their roles in pain regulation.Results: Our results showed that chemogenetic activation in both the RSC and MCC heightened pain sensitivity. Suppression of MCC activity disrupted the RSC's regulation of both mechanical and thermal pain, while RSC inhibition specifically affected the MCC's regulation of thermal pain. The findings indicate a complex interplay between the MCC and RSC, with the MCC potentially governing the RSC's pain regulatory mechanisms. The RSC, in turn, is crucial for the MCC's control over thermal sensation, revealing a collaborative mechanism in pain processing. This study provides evidence for the MCC and RSC's collaborative roles in pain regulation, highlighting the importance of their interactions for thermal and mechanical pain sensitivity. Understanding these mechanisms could aid in developing targeted therapies for pain disorders.

Molecular dynamics analysis of subsurface brittleness mechanism of nanocrystalline 3C-SiC rough friction surface

To study the effect of polycrystalline 3C-SiC rough friction surface on the mechanism of subsurface brittleness during nanocrystalline grinding. Initial grinding models of polycrystalline 3C-SiC and diamond abrasive grains on rough friction surfaces are developed using molecular dynamics methods and the Voronoi method for constructing polycrystalline abrasive grains. The processing mechanism of 3C-SiC is analyzed by post-processing methods such as dislocation defect analysis, atomic arrangement analysis and stress analysis. At 2.6 nm, "stress concentration" occurs between the abrasive particles and the workpiece, forming irregular force shapes. The larger the grain size, the smaller the crystal hardness, the greater the possibility of crystal fracture, and it is obvious in the crystal of larger grains. At 8 nm, the crystal breaks and creates vacancies. The roughness of the polycrystalline 3C-SiC friction surface and the cross-cutting mechanism between grains with grain boundaries are found to be effective in ameliorating the damage in the subsurface layer.

Epigenetic Regulation and Molecular Mechanisms of Burn Injury-Induced Nociception in the Spinal Cord of Mice.

Epigenetic mechanisms, including histone post-translational modifications (PTMs), play a critical role in regulating pain perception and the pathophysiology of burn injury. However, the epigenetic regulation and molecular mechanisms underlying burn injury-induced pain remain insufficiently explored. Spinal dynorphinergic (Pdyn) neurons contribute to heat hyperalgesia induced by severe scalding-type burn injury through p-S10H3-dependent signaling. Beyond p-S10H3, burn injury may impact various other histone H3 PTMs. Double immunofluorescent staining and histone H3 protein analyses demonstrated significant hypermethylation at H3K4me1 and H3K4me3 sites and hyperphosphorylation at S10H3 within the spinal cord. By analyzing Pdyn neurons in the spinal dorsal horn, we found evidence of chromatin activation with a significant elevation in p-S10H3 immunoreactivity. We used RNA-seq analysis to compare the effects of burn injury and formalin-induced inflammatory pain on spinal cord transcriptomic profiles. We identified 98 DEGs for burn injury and 86 DEGs for formalin-induced inflammatory pain. A limited number of shared differentially expressed genes (DEGs) suggest distinct central pain processing mechanisms between burn injury and formalin models. KEGG pathway analysis supported this divergence, with burn injury activating Wnt signaling. This study enhances our understanding of burn injury mechanisms and uncovers converging and diverging pathways in pain models with different origins.

Task-dependent contribution to edge-based versus region-based texture perception

Texture segregation studies indicate that some types of textures are processed by edge-based and others by region-based mechanisms. However, studies employing nominally edge-based textures have found evidence for region-based processing mechanisms when the task was to detect rather than segregate the textures. Here we investigate directly whether the nature of the task determines if region-based or edge-based mechanisms are involved in texture perception. Stimuli consisted of randomly positioned Gabor micropattern texture arrays with five types of modulation: orientation modulation, orientation variance modulation, luminance modulation, contrast modulation and contrast variance modulation (CVM). There were four modulation frequencies: 0.1, 0.2, 0.4 and 0.8 cpd. Each modulation type was defined by three types of waveforms: sinewave (SN) with its smooth variations, square-wave (SQ) and cusp-wave (CS) with its sharp texture edges. The CS waveform was constructed by removing a sinewave from an equal amplitude square-wave. Participants performed two tasks: detection in which participants selected which of two stimuli contained the modulation and discrimination in which participants indicated which of two textures had a different modulation orientation. Our results indicate that threshold amplitudes in the detection task followed the ordering SQ < SN < CS across all spatial frequencies, consistent with detection being mediated by the overall energy in the stimulus and hence region based. With the discrimination task at low texture spatial frequencies and with CVM textures at all spatial frequencies the order was CS ≤ SQ with both < SN, consistent with being edge-based. We modeled the data by estimating the spatial frequency of a Difference of Gaussian filter that gave the largest peak amplitude response to the data. We found that the peak amplitude was lower for detection than discrimination across all texture types except for the CVM texture. We conclude that task requirements are critical to whether edges or regions underpin texture processing.

Perceptual target discriminability modulates the Simon effect beyond the fading of distractor-based activation: Insights from delta plots and diffusion model analyses.

The visual Simon task is widely employed to explore the underlying mechanisms of sensorimotor processing in the presence of task-relevant (targets) and task-irrelevant (distracting) location information. Critically, the Simon effect is considered as an indicator of action-related interference resulting from distractor-based activation, which fades out over time. In this study, we tested whether attenuated Simon effects with slower task processing may be fully explained by the fading of distractor-based response activation. To that end, we selectively manipulated perceptual target discriminability by varying the ratio of differently colored dots within (Experiment 1) and between blocks (Experiment 2). According to pure fading activation accounts, the negative-going delta plots of the two discriminability conditions should overlap across the entire reaction time (RT) distribution. In contrast to this prediction, the negative-going DPs for the two discriminability conditions did not overlap in either experiment. Instead, the Simon effect was either consistently smaller (Experiment 1) or larger (Experiment 2) across the entire RT distribution in the easy condition compared to the hard condition. This result pattern indicates that perceptual target discriminability affected conflict resolution beyond the mere fading of distractor-based activation. Exploratory model-based analyses suggest a stronger processing of relevant perceptual information with more discriminable targets, which may counteract the influence of distracting location information. However, as the exact effects of discriminability on conflict processing seem to depend on variation mode (trialwise vs. blockwise), the importance of global strategic effects is also highlighted. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

A novel motherboard test item yield prediction model based on parallel feature extraction

Abstract Functional testing of motherboards in Surface Mount Technology (SMT) assembly lines is crucial. Accurate yield prediction for each test item optimizes testing strategies, reduces costs, and ensures test coverage. Manual estimation of test item yields remains common, hindering accurate on-site predictions. Existing research on motherboard yield lacks predictions for individual test items and ignores temporal correlations during placement. This paper introduces a method, a convolutional bidirectional long short-term memory attention network (CBA-Net), which combines a convolutional neural network and a bidirectional long short-term memory network with an attention mechanism for parallel processing. It preprocesses historical test data, leveraging both networks to identify key features and extract temporal correlations. The attention mechanism optimizes yield predictions by assigning weights to information at different time steps. Experimental validation using actual production data demonstrates that the proposed method performs better compared to traditional models.

Automated Detection and Management of Leaf Diseases in Plants using Image Processing and Machine Learning

This paper presents a system designed to detect and monitor diseases found on the leaves of plants through the aid of image processing and machine learning mechanisms. The system can diagnose the problem from the images it receives from leaves and recommend specific treatment. Common users are able to front-end an image of the leaf through a simple web interface that feeds the images into a central database. In response to an image query, the system uses the GLCM method to feed texture characteristics in and to employ K - means clustering in segmenting disease regions. After this, a Support Vector Machine (SVM) polynomial kernel-based classifier is used for the multi-class classification of the diseases of the leaves. It also has features that entails information on the diseases identified and treatment plans for them, improving disease control plans. Experimental assessments show that the developed system is truly capable of analyzing the peculiarity of diseased leaves and categorizing the diseases with high prediction accuracy, pointing to a promising application of the proposed system in agricultural industry and plant pathology science. Some possible enhancements for the future could be the expansion of more functionalities or creating a mobile app aimed at enhancing the program’s usage. The proposed system will not only help in moving the agricultural industry forward by providing a dependable, easy to use, error-free system for early diagnosis of diseases and taking initial steps in the mitigation process but also result in improved quality and quantity of produce.

The development of a research intelligence tool for rare disease research in the Netherlands

Abstract This article describes the process of the development of a research intelligence tool to analyze rare disease research in the Netherlands. To the best of our knowledge, this is the first tool that can surface and organize scientific output on rare diseases using established annotation and natural language processing mechanisms. We focus on the track leading up to the development, including strategic motivation and user needs, of a proof-of-concept tool, upscaling the idea to a national collaboration project, the development of the final tool and a usability evaluation and subsequent fine-tuning. The tool is a unique visualization that allows users to benefit with a few clicks from getting the information they require for their needs, and offers novel scientific indicators to characterize (relatively) rare disease research activity. We discuss the applications of insights derived from this tool for science policy and to support decision-making, and to identify opportunities and potential collaborations and make recommendations for future developments, including a broadening of the scope, and discuss potential novel applications.

Antecedents of Risk Information Sharing &amp; Processing Mechanism in Supply Chain: A social and technical Perspective

This paper emphasizes Supply Chain Risk Management (SCRM), whose execution can boost an organization's resilience and operating performance. Drawing on quantitative data from 231 supply chain professionals in various industries, the results highlight that subcomponents of SCRM—specifically strategy alignment and technology usage—significantly influence risk information sharing, analysis, and evaluation. Abstract Surprisingly, an SCRM-oriented culture does not significantly affect the exchange of risk information or the amount of risk partnership. At the same time, team support has insignificant effects on risk analysis and assessment. From these insights, it is evident that although having a culture of SCRM in place and some team involvement is beneficial, they could not, altogether, pave the way toward the organizational capability of evaluating and disseminating risk-related information. In its place, the emphasis on aligned SCRM strategy and technological advancement in the supply chain sector becomes a critical factor for risk mitigation. This suggests an absolute imperative for managers: to facilitate a technological-enabled SCRM approach that is collaborative, in line with the strategy, and allows integrated information processing. Fortifying these domains can help minimize weaknesses, increase response, and streamline the supply chain performance; continuous monitoring and improvement of information dissemination processes are essential, as they facilitate better risk management across the supply chain network.