he advance in Non-Volatile Memory(NVM)has changed the traditional *** to DRAM,NVM has the advantages of nonvolatility and large ***,as the read/write speed of NVM is still lower than that of DRAM,building DRAM/NVM-bas...
详细信息
he advance in Non-Volatile Memory(NVM)has changed the traditional *** to DRAM,NVM has the advantages of nonvolatility and large ***,as the read/write speed of NVM is still lower than that of DRAM,building DRAM/NVM-based hybrid memory systems is a feasible way of adding NVM into the current computer *** paper aims to optimize the well-known B^(+)-tree for hybrid *** novelty of this study is ***,we observed that the space utilization of internal nodes in B^(+)-tree is generally below 70%.Inspired by this observation,we propose to maintain hot keys in the free space within internal nodes,yielding a new index named HATree(Hotness-Aware Tree).The new idea of HATree is to use the unused space of the parent of leaf nodes(PLNs)as the hotspot data ***,no extra space is needed,and the in-node hotspot cache can efficiently improve query ***,to further improve the update performance of HATree,we propose to utilize the eADR technology supported by the third-generation Intel Xeon Scalable Processors to enhance HATree with instant log persistence,which results in the new HATree-Log *** conduct extensive experiments on real hybrid memory architecture involving DRAM and Intel Optane Persistent Memory to evaluate the performance of HATree and *** state-of-the-art indices for hybrid memory,namely NBTree,LBTree,and FPTree,are included in the experiments,and the results suggest the efficiency of HATree and HATree-Log.
In a crowd density estimation dataset,the annotation of crowd locations is an extremely laborious task,and they are not taken into the evaluation *** this paper,we aim to reduce the annotation cost of crowd datasets,a...
详细信息
In a crowd density estimation dataset,the annotation of crowd locations is an extremely laborious task,and they are not taken into the evaluation *** this paper,we aim to reduce the annotation cost of crowd datasets,and propose a crowd density estimation method based on weakly-supervised learning,in the absence of crowd position supervision information,which directly reduces the number of crowds by using the number of pedestrians in the image as the supervised *** this purpose,we design a new training method,which exploits the correlation between global and local image features by incremental learning to train the ***,we design a parent-child network(PC-Net)focusing on the global and local image respectively,and propose a linear feature calibration structure to train the PC-Net simultaneously,and the child network learns feature transfer factors and feature bias weights,and uses the transfer factors and bias weights to linearly feature calibrate the features extracted from the Parent network,to improve the convergence of the network by using local features hidden in the crowd *** addition,we use the pyramid vision transformer as the backbone of the PC-Net to extract crowd features at different levels,and design a global-local feature loss function(L2).We combine it with a crowd counting loss(LC)to enhance the sensitivity of the network to crowd features during the training process,which effectively improves the accuracy of crowd density *** experimental results show that the PC-Net significantly reduces the gap between fullysupervised and weakly-supervised crowd density estimation,and outperforms the comparison methods on five datasets of Shanghai Tech Part A,ShanghaiTech Part B,UCF_CC_50,UCF_QNRF and JHU-CROWD++.
Localizing discriminative object parts(e.g.,bird head)is crucial for fine-grained classification tasks,especially for the more challenging fine-grained few-shot *** work always relies on the learned object parts in a ...
详细信息
Localizing discriminative object parts(e.g.,bird head)is crucial for fine-grained classification tasks,especially for the more challenging fine-grained few-shot *** work always relies on the learned object parts in a unified manner,where they attend the same object parts(even with common attention weights)for different few-shot episodic *** this paper,we propose that it should adaptively capture the task-specific object parts that require attention for each few-shot task,since the parts that can distinguish different tasks are naturally *** for a few-shot task,after obtaining part-level deep features,we learn a task-specific part-based dictionary for both aligning and reweighting part features in an ***,part-level categorical prototypes are generated based on the part features of support data,which are later employed by calculating distances to classify query data for *** retain the discriminative ability of the part-level representations(i.e.,part features and part prototypes),we design an optimal transport solution that also utilizes query data in a transductive way to optimize the aforementioned distance calculation for the final *** experiments on five fine-grained benchmarks show the superiority of our method,especially for the 1-shot setting,gaining 0.12%,8.56%and 5.87%improvements over state-of-the-art methods on CUB,Stanford Dogs,and Stanford Cars,respectively.
All-reduce is a widely used communication technique for distributed and parallel applications typically implemented using either a tree-based or ring-based scheme. Each of these approaches has its own limitations: tre...
详细信息
All-reduce is a widely used communication technique for distributed and parallel applications typically implemented using either a tree-based or ring-based scheme. Each of these approaches has its own limitations: tree-based schemes struggle with efficiently exchanging large messages, while ring-based solutions assume constant communication throughput,an unrealistic expectation in modern network communication infrastructures. We present FMCC-RT, an all-reduce approach that combines the advantages of tree-and ring-based implementations while mitigating their drawbacks. FMCC-RT dynamically switches between tree and ring-based implementations depending on the size of the message being processed. It utilizes an analytical model to assess the impact of message sizes on the achieved throughput, enabling the derivation of optimal work partitioning parameters. Furthermore, FMCC-RT is designed with an Open MPI-compatible API, requiring no modification to user code. We evaluated FMCC-RT through micro-benchmarks and real-world application tests. Experimental results show that FMCC-RT outperforms state-of-the-art tree-and ring-based methods, achieving speedups of up to 5.6×.
In the video captioning methods based on an encoder-decoder,limited visual features are extracted by an encoder,and a natural sentence of the video content is generated using a ***,this kind ofmethod is dependent on a...
详细信息
In the video captioning methods based on an encoder-decoder,limited visual features are extracted by an encoder,and a natural sentence of the video content is generated using a ***,this kind ofmethod is dependent on a single video input source and few visual labels,and there is a problem with semantic alignment between video contents and generated natural sentences,which are not suitable for accurately comprehending and describing the video *** address this issue,this paper proposes a video captioning method by semantic topic-guided ***,a 3D convolutional neural network is utilized to extract the spatiotemporal features of videos during the ***,the semantic topics of video data are extracted using the visual labels retrieved from similar video *** the decoding,a decoder is constructed by combining a novel Enhance-TopK sampling algorithm with a Generative Pre-trained Transformer-2 deep neural network,which decreases the influence of“deviation”in the semantic mapping process between videos and texts by jointly decoding a baseline and semantic topics of video *** this process,the designed Enhance-TopK sampling algorithm can alleviate a long-tail problem by dynamically adjusting the probability distribution of the predicted ***,the experiments are conducted on two publicly used Microsoft Research Video Description andMicrosoft Research-Video to Text *** experimental results demonstrate that the proposed method outperforms several state-of-art ***,the performance indicators Bilingual Evaluation Understudy,Metric for Evaluation of Translation with Explicit Ordering,Recall Oriented Understudy for Gisting Evaluation-longest common subsequence,and Consensus-based Image Description Evaluation of the proposed method are improved by 1.2%,0.1%,0.3%,and 2.4% on the Microsoft Research Video Description dataset,and 0.1%,1.0%,0.1%,and 2.8% on the Microsoft Research-Video to Text dataset
The proliferation of deep learning(DL)has amplified the demand for processing large and complex datasets for tasks such as modeling,classification,and ***,traditional DL methods compromise client privacy by collecting...
详细信息
The proliferation of deep learning(DL)has amplified the demand for processing large and complex datasets for tasks such as modeling,classification,and ***,traditional DL methods compromise client privacy by collecting sensitive data,underscoring the necessity for privacy-preserving solutions like Federated Learning(FL).FL effectively addresses escalating privacy concerns by facilitating collaborative model training without necessitating the sharing of raw *** that FL clients autonomously manage training data,encouraging client engagement is pivotal for successful model *** overcome challenges like unreliable communication and budget constraints,we present ENTIRE,a contract-based dynamic participation incentive mechanism for *** ensures impartial model training by tailoring participation levels and payments to accommodate diverse client *** approach involves several key ***,we examine how random client participation impacts FL convergence in non-convex scenarios,establishing the correlation between client participation levels and model ***,we reframe model performance optimization as an optimal contract design challenge to guide the distribution of rewards among clients with varying participation *** balancing budget considerations with model effectiveness,we craft optimal contracts for different budgetary constraints,prompting clients to disclose their participation preferences and select suitable contracts for contributing to model ***,we conduct a comprehensive experimental evaluation of ENTIRE using three real *** results demonstrate a significant 12.9%enhancement in model performance,validating its adherence to anticipated economic properties.
Multiarmed bandit(MAB) models are widely used for sequential decision-making in uncertain environments, such as resource allocation in computer communication systems.A critical challenge in interactive multiagent syst...
Multiarmed bandit(MAB) models are widely used for sequential decision-making in uncertain environments, such as resource allocation in computer communication systems.A critical challenge in interactive multiagent systems with bandit feedback is to explore and understand the equilibrium state to ensure stable and tractable system performance.
In the wake of rapid advancements in artificial intelligence(AI), we stand on the brink of a transformative leap in data systems. The imminent fusion of AI and DB(AI×DB) promises a new generation of data systems,...
详细信息
In the wake of rapid advancements in artificial intelligence(AI), we stand on the brink of a transformative leap in data systems. The imminent fusion of AI and DB(AI×DB) promises a new generation of data systems, which will relieve the burden on end-users across all industry sectors by featuring AI-enhanced functionalities, such as personalized and automated in-database AI-powered analytics, and selfdriving capabilities for improved system performance. In this paper, we explore the evolution of data systems with a focus on deepening the fusion of AI and DB. We present NeurDB, an AI-powered autonomous data system designed to fully embrace AI design in each major system component and provide in-database AI-powered analytics. We outline the conceptual and architectural overview of NeurDB, discuss its design choices and key components, and report its current development and future plan.
Recommendation has been widely used in business scenarios to provide users with personalized and accurate item lists by efficiently analyzing complex user-item ***,existing recommendation methods have significant shor...
详细信息
Recommendation has been widely used in business scenarios to provide users with personalized and accurate item lists by efficiently analyzing complex user-item ***,existing recommendation methods have significant shortcomings in capturing the dynamic preference changes of users and discovering their true potential *** address these problems,a novel framework named Intent-Aware Graph-Level Embedding Learning(IaGEL)is proposed for *** this framework,the potential user interest is explored by capturing the co-occurrence of items in different periods,and then user interest is further improved based on an adaptive aggregation algorithm,forming generic intents and specific *** addition,for better representing the intents,graph-level embedding learning is designed based on the mutual information comparison among positive intents and negative ***,an intent-based recommendation strategy is designed to further mine the dynamic changes in user *** on three public and industrial datasets demonstrate the effectiveness of the proposed IaGEL in the task of recommendation.
Convertible hydrogel supercapacitors have emerged as promising energy storage devices in switches,diodes,and ***,inherent weaknesses in ionic conductivity,mechanical properties,and water retention of hydrogel electrol...
详细信息
Convertible hydrogel supercapacitors have emerged as promising energy storage devices in switches,diodes,and ***,inherent weaknesses in ionic conductivity,mechanical properties,and water retention of hydrogel electrolytes seriously hinder their *** by the hardness conversion of sea cucumber skin,a conductivity and mechanics dual-tunable salt gel electrolyte is successfully *** salt gel presents a reversible switching of conductors-insulators and a mechanical regulation between softness and hardness via the dissolution-crystallization transition of sodium acetate trihydrate(SAT).Meanwhile,the salt gels spontaneously grow a layer of“armor”through saturated phase-change salt crystals effectively reducing water evaporation of hydrogel ***,this phase-change soft-rigid conversion strategy will expand the capabilities of gel-based flexible supercapacitors(area capacitance:258.6 mF cm^(-2)),and the capacitance retention rate could still reach 86.9%after 3000 cycles at high ***,the salt gel supercapacitor is potentially used in over-heat alarm *** is anticipated that the strategy of conductivity and mechanics of dual-tunable salt gel would provide a new perspective on the development of energy storage devices,wearable electronics,and flexible robots.
暂无评论