We present GauKGT5, a sequence-to-sequence model proposed for knowledge graph completion (KGC). Our research extends the KGT5 model, a recent sequence-to-sequence link prediction (LP) model. GauKGT5 takes advantage of...
详细信息
The extensive development of complex applications in embedded devices has driven the rapid development of edge computing, which provides powerful processing capabilities to the edge network. In this context of develop...
详细信息
Line segment detection is increasingly being widely applied in visual tasks. Traditional methods for line segment detection are known for their speed and accuracy, but they lack robustness in handling noisy images. CN...
详细信息
Spatial join is an important operation for combining spatial data. parallelization is essential for improving spatial join performance. However, load imbalance due to data skew limits the scalability of parallel spati...
详细信息
ISBN:
(纸本)9781450395298
Spatial join is an important operation for combining spatial data. parallelization is essential for improving spatial join performance. However, load imbalance due to data skew limits the scalability of parallel spatial join. There are many work sharing techniques to address this problem in a parallel environment. One of the techniques is to use data and space partitioning and then scheduling the partitions among threads/processes with the goal of minimizing workload differences across threads/processes. However, load imbalance still exists due to differences in join costs of different pairs of input geometries in the partitions. For the load imbalance problem, we have designed a work stealing spatial join system (WSSJ-DM) on a distributed memory environment. Work stealing is an approach for dynamic load balancing in which an idle processor steals computational tasks from other processors [5]. This is the first work that uses work stealing concept (instead of work sharing) to parallelize spatial join computation on a large compute cluster. We have evaluated the scalability of the system on shared and distributed memory. Our experimental evaluation shows that work stealing is an effective strategy. We compared WSSJ-DM with work sharing implementations of spatial join on a high performance computing environment using partitioned and un-partitioned datasets. Static and dynamic load balancing approaches were used for comparison. We study the effect of memory affinity in work stealing operations involved in spatial join on a multi-core processor. WSSJ-DM performed spatial join using ST_Intersection on Lakes (8.4M polygons) and Parks (10M polygons) in 30 seconds using 35 compute nodes on a cluster (1260 CPU cores). A work sharing Master-Worker implementation took 160 seconds in contrast.
Users' behaviours show a noticeable impact on cloud computing resources. Behaviour prediction models could foster usage awareness of cloud users. This requires training prediction models with datasets that provide...
详细信息
ISBN:
(纸本)9781665469586
Users' behaviours show a noticeable impact on cloud computing resources. Behaviour prediction models could foster usage awareness of cloud users. This requires training prediction models with datasets that provide user information. Unfortunately, such information is excluded from many relevant datasets. Therefore, in this work, we investigate the ability of extracting these identities via clustering methods. We conduct this by categorising workload datasets according to the availability of users information in their attributes. Then, we focus our attention on shared attributes between user information disclosing and non-disclosing datasets. Eventually, we evaluated the potential of several clustering approaches on user information disclosing datasets. Our results show that users' identifications can be extracted with relatively high accuracy using clustering. They also show that the highest clustering precision is mostly obtained from the attributes representing request components that strongly relate to the user's application.
This paper proposes a scalable and efficient architecture to accelerate random forest computation on FPGA devices targeting edge computing platforms. The proposed architecture with efficient decision tree units (DTUs)...
详细信息
ISBN:
(纸本)9783031506833;9783031506840
This paper proposes a scalable and efficient architecture to accelerate random forest computation on FPGA devices targeting edge computing platforms. The proposed architecture with efficient decision tree units (DTUs) executes samples in a pipeline model for improving performance. Moreover, a size-effective memory organization is also introduced with the architecture to save the on-chip block ram used for reducing the latency and improving working frequency of the implementation system on FPGA devices. We target edge computing platforms that suffer from the limitations of resources and power consumption. Therefore, the proposed architecture can reconfigure the number of DTUs according to the target platform's available resources. We build a system with a PYNQ Z2 FPGA board for testing, validating, and estimating the proposed architecture. In this system, we exploit different numbers of DTUs, from 1 to 15, to test our scalability. Experimental results with certified datasets show that we achieve speed-ups by up to 170.39x and 90.27x compared to Intel core i7 desktop version and core i9 high-performance computing version processors, respectively.
Vehicular edge computing (VEC) as a promising computing paradigm has accelerated the reformation of existing dominating computing infrastructures, enabling resource provisioning in close proximity to resource requesto...
详细信息
With recent advances in deep learning and large-scale computing, learning-based controls have become increasingly attractive for complex physical systems. Consequently, developing generalized learning-based control so...
详细信息
ISBN:
(纸本)9798400702273
With recent advances in deep learning and large-scale computing, learning-based controls have become increasingly attractive for complex physical systems. Consequently, developing generalized learning-based control software that takes into account the next generation of computing architectures is paramount. Specifically, for the case of complex control, we present the Easily eXtendable Architecture for Reinforcement Learning (EXARL), which aims to support various scientific applications seeking to leverage reinforcement learning (RL) on exascale computing architectures. We demonstrate the efficacy and performance of the EXARL library for the scientific use case of designing a complex control policy to stabilize a power system after experiencing a fault. We use a parallel augmented random search method developed within EXARL and present its preliminary validation and performance stabilization of a fault for the IEEE 39-bus system.
In the current global energy crisis, distributed photovoltaic power generation represents an innovative approach to integrated power generation and energy utilization, offering significant prospects for development. H...
详细信息
Federated Learning, a disruptive and novel aspect of machine learning, is at the forefront of decentralized, privacy-conscious data processing. This in-depth review study navigates the complex environment of Federated...
详细信息
暂无评论