This paper considers improving wireless communication and computation efficiency in federated learning (FL) via model quantization. In the proposed bitwidth FL scheme, edge devices train and transmit quantized version...
详细信息
ISBN:
(纸本)9781665435413
This paper considers improving wireless communication and computation efficiency in federated learning (FL) via model quantization. In the proposed bitwidth FL scheme, edge devices train and transmit quantized versions of their local FL model parameters to a coordinating server, which, in turn, aggregates them into a quantized global model and synchronizes the devices. With the goal of jointly determining the set of participating devices in each training iteration and the bitwidths employed at the devices, we pose an optimization problem for minimizing the training loss of quantized FL under a device sampling budget and delay requirement. Our analytical results show that the improvement of FL training loss between two consecutive iterations depends on not only the device selection and quantization scheme, but also on several parameters inherent to the model being learned. As a result, we propose, a model-based reinforcement learning (RL) method to optimize action selection over iterations. Compared to model-free RL, the proposed approach leverages the derived mathematical characterization of the FL training process to discover an effective device selection and quantization scheme without imposing additional device communication overhead. Numerical evaluations show that the proposed FL framework can achieve the same classification performance while reducing the number of training iterations needed for convergence by 20% compared to model-free RL-based FL.
Medical report generation (MRG) is essential for computer-aided diagnosis and medication guidance, which can relieve the heavy burden of radiologists by automatically generating the corresponding medical reports accor...
详细信息
To classify the X-ray mammograms images as benign or malignant is a long-standing unresolved problem, due to the high similarity of different between the mammograms images. In this study, a novel convolutional neural ...
详细信息
In this study, computer simulations are performed on three-dimensional granular systems under shear conditions. The system comprises granular particles that are confined between two rigid plates. The top plate is subj...
详细信息
In this study, computer simulations are performed on three-dimensional granular systems under shear conditions. The system comprises granular particles that are confined between two rigid plates. The top plate is subjected to a normal force and driven by a shearing velocity. A positive shear-rate dependence of granular friction, known as velocity-strengthening, exists between the granular and shearing plate. To understand the origin of the dependence of frictional sliding, we treat the granular system as a complex network, where granular particles are nodes and normal contact forces are weighted edges used to obtain insight into the interiors of granular matter. Community structures within granular property networks are detected under different shearing velocities in the steady state. Community parameters, such as the size of the largest cluster and average size of clusters, show significant monotonous trends in shearing velocity associated with the shear-rate dependence of granular friction. Then, we apply an instantaneous change in shearing velocity. A dramatic increase in friction is observed with a change in shearing velocity in the non-steady state. The community structures in the non-steady state are different from those in the steady state. Results indicate that the largest cluster is a key factor affecting the friction between the granular and shearing plate.
In this paper, we generalize the concept of strong quantum nonlocality from two aspects. Firstly in Cd ⊗ Cd ⊗ Cd quantum system, we present a construction of strongly nonlocal quantum states containing 6(d−1)2 orthogo...
详细信息
Pushing artificial intelligence (AI) from central cloud to network edge has reached board consensus in both industry and academia for materializing the vision of artificial intelligence of things (AIoT) in the sixth-g...
详细信息
We probe the spectrum of elementary excitations in SrIrO3 by using heterostructured [(SrIrO3)m/(SrTiO3)l] samples to approach the bulk limit. Our resonant inelastic x-ray scattering (RIXS) measurements at the Ir L3-ed...
详细信息
Image paragraph captioning (IPC) aims to generate a fine-grained paragraph to describe the visual content of an image. Significant progress has been made by deep neural networks, in which the attention mechanism plays...
详细信息
This paper defines a new visual reasoning paradigm by introducing an important factor, i.e. transformation. The motivation comes from the fact that most existing visual reasoning tasks, such as CLEVR in VQA, are solel...
详细信息
Federated learning (FL), as a promising distributed learning paradigm, has put many efforts into distributed intrusion detection systems (IDS), for defending against various malicious attacks, such as SQL injection an...
详细信息
ISBN:
(纸本)9781728181059
Federated learning (FL), as a promising distributed learning paradigm, has put many efforts into distributed intrusion detection systems (IDS), for defending against various malicious attacks, such as SQL injection and DDoS attacks. Compared with traditional IDS based on centralized deep learning (DL), FL-based solutions require not to share users' raw data while yielding better detection performance. However, state-of-the-art FL-based methods still suffer from two key limitations: 1) insufficient detection performance on non-independent and identically distributed (non-IID) data, and 2) high communication and computational overheads due to the utilization of large-scale neural network models. In this paper, we propose a lightweight collaborative intrusion detection framework, called CoLGBM, the first of its kind in the regime of decentralized IDS, where decision tree and light gradient boosting machine (LGBM) are combined for constructing the detection scheme. The main insight is that through combining user-trained decision trees (each user's decision tree is derived from its own data with unique distribution), our framework can perform effectively on non-IID data while working efficiently for handling enormous samples. Compared with the current FL-based methods, our CoLGBM achieves higher accuracy and lower overhead on both IID and non-IID data. Extensive experiment results demonstrate our scheme with high-level performance.
暂无评论