A floating horizontal-axis tidal current turbine(HATT)is an underwater power generation device where cavitation inevitably occurs on blade surfaces,severely affecting a turbine’s *** wave action,these floating turbin...
详细信息
A floating horizontal-axis tidal current turbine(HATT)is an underwater power generation device where cavitation inevitably occurs on blade surfaces,severely affecting a turbine’s *** wave action,these floating turbines exhibit six degrees of freedom motion,potentially intensifying the cavitation on the blade *** study selects three types of oscillatory motions from the six degrees of freedom:roll,yaw,and *** fluid dynamics(CFD)methods are used for numerical calculations,and transient simulations of multiphase flow are conducted on the basis of the Reynolds-Averaged Navier-Stokes(RANS)*** has revealed strong correlations between flow velocity,the blade tip speed ratio,and *** oscillatory motion,the motion period and amplitude also significantly impact *** roll motion,the cavitation rate can increase by up to 59%with decreasing period,whereas in pitch and yaw motions,the increases are 7.57 times and 36%larger,*** an increase in amplitude during roll motion,the cavitation rate can increase by up to 1.08 times,whereas in pitch and yaw motions,the increases are 3.49 times and 45%,*** cavitation rate on the blade surfaces is the highest in pitch motion,followed by roll and yaw motions.
We have witnessed the emergence of superhuman intelligence thanks to the fast development of large language models(LLMs) and multimodal language models. As the application of such superhuman models becomes increasingl...
详细信息
We have witnessed the emergence of superhuman intelligence thanks to the fast development of large language models(LLMs) and multimodal language models. As the application of such superhuman models becomes increasingly popular, a critical question arises: how can we ensure they still remain safe, reliable, and aligned well with human values encompassing moral values, Schwartz's Values, ethics, and many more? In this position paper, we discuss the concept of superalignment from a learning perspective to answer this question by outlining the learning paradigm shift from large-scale pretraining and supervised fine-tuning, to alignment training. We define superalignment as designing effective and efficient alignment algorithms to learn from noisy-labeled data(point-wise samples or pair-wise preference data) in a scalable way when the task is very complex for human experts to annotate and when the model is stronger than human experts. We highlight some key research problems in superalignment, namely, weak-to-strong generalization, scalable oversight, and evaluation. We then present a conceptual framework for superalignment, which comprises three modules: an attacker which generates the adversary queries trying to expose the weaknesses of a learner model, a learner which refines itself by learning from scalable feedbacks generated by a critic model with minimal human experts, and a critic which generates critics or explanations for a given query-response pair, with a target of improving the learner by criticizing. We discuss some important research problems in each component of this framework and highlight some interesting research ideas that are closely related to our proposed framework, for instance, self-alignment, self-play, self-refinement, and more. Last, we highlight some future research directions for superalignment, including the identification of new emergent risks and multi-dimensional alignment.
This article designs the PELAN structure based on the lightweight YOLOv7-tiny model for surface defect detection of hot-rolled steel strips. At the same time, the CA (Channel Attention) is embedded in the feature pyra...
详细信息
Iris biometrics allow contactless authentication, which makes it widely deployed human recognition mechanisms since the couple of years. Susceptibility of iris identification systems remains a challenging task due to ...
详细信息
We examined the enhancing effects of different dosages of product of Centrifugation of Bacterial Liquid(product of CBL)on the performance of slag-fGD gypsum-cement-bentonite-sludge system using MICP *** analyzed the m...
详细信息
We examined the enhancing effects of different dosages of product of Centrifugation of Bacterial Liquid(product of CBL)on the performance of slag-fGD gypsum-cement-bentonite-sludge system using MICP *** analyzed the multifaceted performance of the solidified sludge from macroscopic and microscopic *** experimental results reveal that the increase in product of CBL dosage results in positive impacts on the solidified sludge,including higher side compressive strength,lower leachate heavy metal concentration,and improved crack repair *** a 0.4%product of CBL doping concentration,the strength of the solidified sludge is enhanced by 26.6%at 3 d,61.2%at 7 d,and 13.9%at 28 d when compared to the unmodified solidified *** 28 days,the concentrations of Zn and Cu ions reduce by 58%and 18%,respectively,and the crack repair rate is 58.4%.These results demonstrate that the increase in heavy metal concentration in the leachate leads to an increase in the strength of the solidified *** strengthening procedure heavily relies on the mineralisation reaction of Bacillus pasteurii,which produces a substantial amount of CaCO_(3)to cement the particles and fill the pores *** modified solidifying sludge exhibits a self-repairing effect and an enhanced multifaceted performance as a result of oxygen being restored after crack formation and reactivation of Bacillus *** conditions facilitate the body's recovery.
Device-to-device(D2D)communication underlay cellular networks offers several benefits,including offloading cellular traffic and improving *** this paper,a scheme is proposed for D2D communication underlay cellular net...
详细信息
Device-to-device(D2D)communication underlay cellular networks offers several benefits,including offloading cellular traffic and improving *** this paper,a scheme is proposed for D2D communication underlay cellular networks,by taking resource allocation algorithm into ***,a scenario where cellular users(CUs)and D2D users(DUs)coexist is *** characterize the clustering feature of DUs,the locations of CUs and DUs are modeled as Poisson cluster processes(PCPs),and the locations of base stations(BSs)are modeled as a Poisson point process(PPP).Subsequently,a resource allocation algorithm is proposed to allocate resources for CUs to DUs in the cluster,which involves the optimization of matching between D2D user and cellular ***,the coverage probability of the clustering network is *** simulation results indicate that the proposed scheme can effectively enhance system capacity and reveal the effects of various parameters on system performance.
The Hungarian algorithm is a well-known cubic-time algorithm for finding minimum-cost matchings in weighted bipartite graphs. While utilizing it for multi-agent path planning yields the minimum-total-length set of pat...
详细信息
Diabetic Retinopathy (DR) is a common and significant complication in patients with diabetes, and severely affecting their quality of life. Image segmentation plays a crucial role in the early diagnosis and treatment ...
详细信息
Predicting students’academic achievements is an essential issue in education,which can benefit many stakeholders,for instance,students,teachers,managers,*** with online courses such asMOOCs,students’academicrelatedd...
详细信息
Predicting students’academic achievements is an essential issue in education,which can benefit many stakeholders,for instance,students,teachers,managers,*** with online courses such asMOOCs,students’academicrelateddata in the face-to-face physical teaching environment is usually sparsity,and the sample size is *** makes building models to predict students’performance accurately in such an environment even *** paper proposes a Two-WayNeuralNetwork(TWNN)model based on the bidirectional recurrentneural network and graph neural network to predict students’next semester’s course performance using only theirprevious course *** experiments on a real dataset show that our model performs better thanthe baselines in many indicators.
Plug-in Hybrid Electric Vehicles(PHEVs)represent an innovative breed of transportation,harnessing diverse power sources for enhanced *** management strategies(EMSs)that coordinate and control different energy sources ...
详细信息
Plug-in Hybrid Electric Vehicles(PHEVs)represent an innovative breed of transportation,harnessing diverse power sources for enhanced *** management strategies(EMSs)that coordinate and control different energy sources is a critical component of PHEV control technology,directly impacting overall vehicle *** study proposes an improved deep reinforcement learning(DRL)-based EMSthat optimizes realtime energy allocation and coordinates the operation of multiple power *** DRL algorithms struggle to effectively explore all possible state-action combinations within high-dimensional state and action *** often fail to strike an optimal balance between exploration and exploitation,and their assumption of a static environment limits their ability to adapt to changing ***,these algorithms suffer from low sample ***,these factors contribute to convergence difficulties,low learning efficiency,and *** address these challenges,the Deep Deterministic Policy Gradient(DDPG)algorithm is enhanced using entropy regularization and a summation tree-based Prioritized Experience Replay(PER)method,aiming to improve exploration performance and learning efficiency from experience ***,the correspondingMarkovDecision Process(MDP)is ***,an EMSbased on the improvedDRLmodel is *** simulation experiments are conducted against rule-based,optimization-based,andDRL-based *** proposed strategy exhibitsminimal deviation fromthe optimal solution obtained by the dynamic programming(DP)strategy that requires global *** the typical driving scenarios based onWorld Light Vehicle Test Cycle(WLTC)and New European Driving Cycle(NEDC),the proposed method achieved a fuel consumption of 2698.65 g and an Equivalent Fuel Consumption(EFC)of 2696.77 *** to the DP strategy baseline,the proposed method improved the fuel efficiency variances(FEV)by 18.13%
暂无评论