In this paper we consider the problem of training a Support Vector Machine (SVM) online using a stream of data in random order. We provide a fast online training algorithm for general SVM on very large datasets. Based...
详细信息
The intensification of urbanization has exacerbated traffic congestion issues, necessitating intelligent traffic light control system to optimize traffic flow. This study designs and implements an intelligent traffic ...
详细信息
Whilst Convolutional Neural Network (CNN)-based object tracking methods can achieve promising results on traditional well-lit datasets, it is challenging to accurately locate targets in low-light images taken in night...
详细信息
Graph neural networks (GNNs) have become the state of the art for various graph-related tasks and are particularly prominent in heterogeneous graphs (HetGs). However, several issues plague this paradigm: first, the di...
详细信息
The brain is probably the most complex organ in the human body. To understand processes such as learning or healing after brain lesions, we need suitable tools for brain simulations. The Model of Structural Plasticity...
详细信息
ISBN:
(纸本)9783031125973;9783031125966
The brain is probably the most complex organ in the human body. To understand processes such as learning or healing after brain lesions, we need suitable tools for brain simulations. The Model of Structural Plasticity offers a solution to that problem. It provides a way to model the brain bottom-up by specifying the behavior of the neurons and using structural plasticity to form the synapses. However, its original formulation involves a pairwise evaluation of attraction kernels, which drastically limits scalability. While this complexity has recently been decreased to O(***(2)n) after reformulating the task as a variant of an nbody problem and solving it using an adapted version of the Barnes-Hut approximation, we propose an even faster approximation based on the fast multipole method (FMM). The fast multipole method was initially introduced to solve pairwise interactions in linear time. Our adaptation achieves this time complexity, and it is also faster in practice than the previous approximation.
How ions evolve in the Earth’s ion foreshock is a basic problem in the heliosphere community,and the ion beam instability is usually proposed to be one major mechanism affecting the ion dynamics *** work will perform...
详细信息
How ions evolve in the Earth’s ion foreshock is a basic problem in the heliosphere community,and the ion beam instability is usually proposed to be one major mechanism affecting the ion dynamics *** work will perform comprehensive analyses of the oblique ion beam instability in the Earth’s ion *** show that in addition to two well-known parallel instabilities (i.e.,the parallel fast-magnetosonic whistler instability and the parallel Alfvén ion cyclotron instability),the oblique Alfvén ion beam (OA/IB) instability can also be triggered by free energy relating to the relative drift d V between the solar wind proton and reflected proton *** slow d V (e.g.,d V■2.2V_(A),where VAdenotes the Alfvén speed),it only triggers the OA/IB *** d V■2.2V_(A),the growth rate in the OA/IB instability can be about 0.6 times the maximum growth rate in parallel ***,this work finds the existence of two types of OA/IB *** first one appears at slow d V and in the small wavenumber region at fast d V,and this instability can be described by the cold fluid *** second one arises in large wavenumber regions at fast d V,and this instability only appears in warm ***,through the energy transfer rate method,we propose that the OA/IB instability is driven by the competition among the Landau and cyclotron wave-particle interactions of beam protons,the cyclotron wave-particle interaction of core protons,and the Landau wave-particle interaction of *** oblique waves can experience significant damping,the importance of the OA/IB instability may be the effective heating of ions in the Earth’s foreshock.
The performance evaluation of candidate architectures is a key step in evolution-based neural architecture search (ENAS). Generally, high-fidelity evaluation is desired for finding the optimal architecture but suffers...
详细信息
Research into automatically searching for an optimal neural network(NN)by optimi-sation algorithms is a significant research topic in deep learning and artificial ***,this is still challenging due to two issues:Both t...
详细信息
Research into automatically searching for an optimal neural network(NN)by optimi-sation algorithms is a significant research topic in deep learning and artificial ***,this is still challenging due to two issues:Both the hyperparameter and ar-chitecture should be optimised and the optimisation process is computationally *** tackle these two issues,this paper focusses on solving the hyperparameter and architecture optimization problem for the NN and proposes a novel light‐weight scale‐adaptive fitness evaluation‐based particle swarm optimisation(SAFE‐PSO)***,the SAFE‐PSO algorithm considers the hyperparameters and architectures together in the optimisation problem and therefore can find their optimal combination for the globally best ***,the computational cost can be reduced by using multi‐scale accuracy evaluation methods to evaluate ***,a stagnation‐based switch strategy is proposed to adaptively switch different evaluation methods to better balance the search performance and computational *** SAFE‐PSO algorithm is tested on two widely used datasets:The 10‐category(i.e.,CIFAR10)and the 100−cate-gory(i.e.,CIFAR100).The experimental results show that SAFE‐PSO is very effective and efficient,which can not only find a promising NN automatically but also find a better NN than compared algorithms at the same computational cost.
Large language models (LLMs) have rapidly advanced and demonstrated impressive capabilities. In-Context Learning (ICL) and Parameter-Efficient Fine-Tuning (PEFT) are currently two mainstream methods for augmenting LLM...
Stochastic gradient descent(SGD)-based optimizers play a key role in most deep learning models,yet the learning dynamics of the complex model remain obscure. SGD is the basic tool to optimize model parameters, and is ...
详细信息
Stochastic gradient descent(SGD)-based optimizers play a key role in most deep learning models,yet the learning dynamics of the complex model remain obscure. SGD is the basic tool to optimize model parameters, and is improved in many derived forms including SGD momentum and Nesterov accelerated gradient(NAG). However, the learning dynamics of optimizer parameters have seldom been studied. We propose to understand the model dynamics from the perspective of control theory. We use the status transfer function to approximate parameter dynamics for different optimizers as the first-or second-order control system, thus explaining how the parameters theoretically affect the stability and convergence time of deep learning models, and verify our findings by numerical experiments.
暂无评论