查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Mosquito-Borne Disease s-Malaria is the subject of a report. According to news reporting out of Morog oro, Tanzania, by NewsRx editors, research stated, "Accurately determining the a ge and survival probabilities of adult mosquitoes is crucial for understanding p arasite transmission, evaluating the effectiveness of control interventions and assessing disease risk in communities. This study was aimed at demonstrating the rapid identification of epidemiologically relevant age categories of Anopheles funestus, a major Afro-tropical malaria vector, through the innovative combinati on of infrared spectroscopy and machine learning, instead of the cumbersome prac tice of dissecting mosquito ovaries to estimate age based on parity status." Financial supporters for this research include Medical Research Council, Wellcom e Trust, Academy Medical Sciences Springboard Award, Bill and Melinda Gates Foun dation, Royal Society, Howard Hughes Medical Institute.
查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Research findings on Machine Learning are discussed in a new report. According to news reporting originating in Brauns chweig, Germany, by NewsRx journalists, research stated, "This work presents a m achine learning approach to optimize the energy efficiency (EE) in a multi-cell wireless network. This optimization problem is non-convex and its global optimum is difficult to find." Financial support for this research came from Federal Ministry of Education & Research (BMBF). The news reporters obtained a quote from the research from Technical University Braunschweig (TU Braunschweig), "In the literature, either simple but suboptimal approaches or optimal methods with high complexity are proposed. In contrast, w e propose an unsupervised machine learning framework to approach the global opti mum. While the neural network (NN) training takes moderate time, application wit h the trained model requires very low computational complexity. In particular, w e introduce a novel objective function based on stochastic actions to solve the non-convex optimization problem. Besides, we design a dedicated NN architecture SINRnet for the power allocation problems in the interference channel that is pe rmutation-equivariant. We encode our domain knowledge into the NN design and she d light into the black box of machine learning. Training and testing results sho w that the proposed method without supervision and with reasonable computational effort achieves an EE close to the global optimum found by the branch-and-bound algorithm and outperform the successive convex approximation (SCA) algorithm." According to the news reporters, the research concluded: "Hence, the proposed ap proach balances between computational complexity and performance."
查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Research findings on Machine Learning-Artificial Intelligence are discussed in a new report. According to news repor ting originating from Chengdu, People's Republic of China, by NewsRx corresponde nts, research stated, "A simple but effective channel attention module is propos ed for Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR). The ch annel attention technique has shown recent success in improving Deep Convolution al Neural Networks (CNN)." Financial support for this research came from National Natural Science Foundatio n of China (NSFC). Our news editors obtained a quote from the research from the University of Elect ronic Science and Technology of China, "The resolution of SAR images does not su rpass optical images thus information flow of SAR images becomes relatively poor when the network depth is raised blindly leading to a serious gradients explosi on/vanishing. To resolve the issue of SAR image recognition efficiency and ambig uity trade-off, we proposed a simple Channel Attention module into the ResNet Ar chitecture as our network backbone, which utilizes few parameters yet results in a performance gain. Our simple attention module, which follows the implementati on of Efficient Channel Attention, shows that avoiding dimensionality reduction is essential for learning as well as an appropriate cross-channel interaction ca n preserve performance and decrease model complexity. We also explored the One P olicy Learning Rate on the ResNet-50 architecture and compared it with the propo sed attention based ResNet-50 architecture. A thorough analysis of the MSTAR Dat aset demonstrates the efficacy of the suggested strategy over the most recent fi ndings."
查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Researchers detail new data in Machine Learning. According to news reporting originating in Mawson Lakes, Australia, b y NewsRx journalists, research stated, "This study investigated the influence of input parameters on the shear strength of RC squat walls using machine learning (ML) models and finite element method (FEM) analysis. The analyses were conduct ed on the largest currently available dataset of 639 squat RC walls with a heigh t-to-length ratio of less than or equal to 2.0." The news reporters obtained a quote from the research from the University of Sou th Australia, "The findings suggest that ensemble learning models, specifically XGBoost, CatBoost, GBRT, and RF, are effective in predicting the shear strength of RC short shear walls and using Bayesian Optimization for hyperparameter tunin g improves their performance. The axial load had a greater influence on the shea r strength than reinforcement ratio, and longitudinal reinforcement had a more s ignificant impact compared to horizontal and vertical reinforcement. The perform ance of XGBoost model significantly outperforms traditional design models such a s ACI 318-19, ASCE/SEI 43-05, and Wood 1990. Additionally, reducing the number o f input features from 13 to 10, 8, or 6 still yields reliable predictions with h igh accuracy. The finding suggests that the use of XGBoost models provides not o nly comparable accuracy to FEM simulations with non-linear pushover analysis but also the first one can predict the lateral strength in the case of incomplete d ata which could not be done by FEM."
查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Robotics is the subjec t of a report. According to news reporting out of Aarhus, Denmark, by NewsRx edi tors, research stated, "RNA nanotechnology aims to use RNA as a programmable mat erial to create self-assembling nanodevices for application in medicine and synt hetic biology. The main challenge is to develop advanced RNA robotic devices tha t both sense, compute, and actuate to obtain enhanced control over molecular pro cesses." Our news journalists obtained a quote from the research from Aarhus University, "Here, we use the RNA origami method to prototype an RNA robotic device, named t he ‘Traptamer,' that mechanically traps the fluorescent aptamer, iSpinach. The T raptamer is shown to sense two RNA key strands, acts as a Boolean AND gate, and reversibly controls the fluorescence of the iSpinach aptamer. Cryo-electron micr oscopy of the closed Traptamer structure at 5.45-angstrom resolution reveals the mechanical mode of distortion of the iSpinach motif." According to the news editors, the research concluded: "Our study suggests a gen eral approach to distorting RNA motifs and a path forward to build sophisticated RNA machines that through sensing, computing, and actuation modules can be used to precisely control RNA functionalities in cellular systems." For more information on this research see: An RNA origami robot that traps and r eleases a fluorescent aptamer. Science Advances, 2024;10(12). Science Advances can be contacted at: Amer Assoc Advancement Science, 1200 New York Ave, NW, Was hington, DC 20005, USA.
查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Research findings on Robotics-Roboti cs and Automation are discussed in a new report. According to news reporting ori ginating from Kanpur, India, by NewsRx correspondents, research stated, "In this work, we propose an end-to-end Thrust Microstepping and Decoupled Control (TMDC ) of quadrotors. TMDC focuses on precise off-centered aerial grasping of payload s dynamically, which are attached rigidly to the UAV body via a gripper contrary to the swinging payload." Our news editors obtained a quote from the research from Indian Institute for Te chnology, "The dynamic payload grasping quickly changes UAV's mass, inertia etc, causing instability while performing a grasping operation in-air. We identify t hat to handle unknown payload grasping, the role of thrust controller is crucial . Hence, we focus on thrust control without involving system parameters such as mass etc. TMDC is based on our novel Thrust Microstepping via Acceleration Feedb ack (TMAF) thrust controller and Decoupled Motion Control (DMC). TMAF precisely estimates the desired thrust even at smaller loop rates while DMC decouples the horizontal and vertical motion to counteract disturbances in the case of dynamic payloads. We prove the controller's efficacy via exhaustive experiments in prac tically interesting and adverse real-world cases, such as fully onboard state es timation without any positioning sensor, narrow and indoor flying workspaces wit h intense wind turbulence, heavy payloads, non-uniform loop rates, etc."
查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-A new study on pattern recognition and artificial intelligence is now available. According to news originating from Na njing, People's Republic of China, by NewsRx correspondents, research stated, "T he main challenge for object detection in aerial images is small object detectio n." Funders for this research include National Natural Science Foundation of China. Our news reporters obtained a quote from the research from Nanjing University: " Most existing methods use feature fusion strategies to enhance small object feat ures in shallow layers but ignore the problem of inconsistent small object local region responses between feature layers, namely the semantic gap, which may lea d to underutilization of small object information in multiple feature layers. To lift the above limitations, we propose a scale enhancement module that adaptive ly passes valuable small object features in different feature layers to shallow layers to alleviate the semantic gap problem. In particular, the module includes the novel fine-coarse self-attention mechanism, which captures global contextua l information by performing strong interaction of pixel-level information at the local scale and weak interaction of regionlevel information at the global scal e. In addition, the anchor assignment strategy based on the Intersection over Un ion (IoU) metric is not favorable for small objects as the IoU metric for small objects has a lower tolerance for position deviation compared to large ones."
查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News-New research on artificial intelligence is the su bject of a new report. According to news reporting out of Brisbane, Australia, b y NewsRx editors, research stated, "Throughout history, natural disasters have c aused severe damage to people and properties worldwide." The news journalists obtained a quote from the research from Queensland Universi ty of Technology: "Flooding is one of the most disastrous types of natural disas ters. A key feature of flood assessment has been making use of the information d erived from remote-sensing imagery from optical sensors on satellites using spec tral indices. Here, a study was conducted about a recent spectral index, the Nor malised Difference Inundation Index, and a new ensemble spectral index, the Conc atenated Normalised Difference Water Index, and two mature spectral indices: Nor malised Difference Water Index and the differential Normalised Difference Water Index with four different machine learning algorithms: Decision Tree, Random For est, Naive Bayes, and K-Nearest Neighbours applied to the PlanetScope satellite imagery about the Brisbane February 2022 flood which is in urban environment. St atistical analysis was applied to evaluate the results. Overall, the four algori thms provided no significant difference in terms of accuracy and F1 score."
查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Current study results on Machine Learn ing-Artificial Intelligence have been published. According to news reporting o ut of Msila, Algeria, by NewsRx editors, the research stated, "Probabilistic dat abases have emerged as an extension of relational databases that can handle unce rtain data under possible worlds semantics. Although the problems of creating ef fective means of probabilistic data representation as well as probabilistic quer y evaluation have been addressed so widely, low attention has been given to quer y result explanation." Our news journalists obtained a quote from the research from University Mohamed Boudiaf-M'sila, "While query answer explanation in relational databases tends to answer the question: why is this tuple in the query result? In probabilistic da tabases, we should ask an additional question: why does this tuple have such a p robability? Due to the huge number of resulting worlds of probabilistic database s, query explanation in probabilistic databases is a challenging task. In this p aper, we propose a causal explanation technique for conjunctive queries in proba bilistic databases. Based on the notions of causality, responsibility and blame, we will be able to address explanation for tuple and attribute uncertainties in a complementary way. Through an experiment on the real-dataset of IMDB, we will see that this framework would be helpful for explaining complex queries results ."
查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Researchers detail new data in Robotic s-Field Robotics. According to news reporting from Shenyang, People's Republic of China, by NewsRx journalists, research stated, "The low-velocity penetrator (LVP) is a planetary penetration device that can drive itself to a target depth through its internal periodic impacts. When LVP generates impact energy, it inev itably produces a recoil that can only be counteracted by friction with the soil , if there is no other auxiliary device." Funders for this research include National Key R&D Program of China , National Natural Science Foundation of China (NSFC), CAS Interdisciplinary Inn ovation Team. The news correspondents obtained a quote from the research from the Chinese Acad emy of Sciences, "Unfortunately, LVP is extraordinarily sensitive to the recoil during the initial stage since the small contact area with the soil results in m inor friction between them. Significantly, once the recoil exceeds the friction, LVP cannot work properly and may even retreat, inducing mission failure. In thi s paper, we develop an optimized LVP with an auxiliary device for lower recoil a nd higher performance. Specifically, we establish a dynamic model to analyze the single-cycle motion of LVP and provide essential support for its optimization a nd design. Meanwhile, an integration method is proposed to calculate the frictio n between LVP and the soil reasonably and accurately. On the basis of these, we obtain the optimal mass and stiffness parameters of LVP that meet both high pene tration efficiency and low recoil. Furthermore, only relying on the parameter op timization is insufficient to eliminate the recoil, and an auxiliary penetration scheme is proposed to provide an external force counteracting the recoil until LVP arrives at a certain depth. Through multiple comparative penetration experim ents, we validate the effectiveness of our approaches in promoting penetration a bility, stability, and restraining the recoil of LVP."
NETL
NSTL