Federated learning is widely accepted as a privacy-preserving paradigm for training a shared global model across multiple client devices in a collaborative fashion. However, in practice, the significantly limited comp...
详细信息
Federated learning is widely accepted as a privacy-preserving paradigm for training a shared global model across multiple client devices in a collaborative fashion. However, in practice, the significantly limited computational power on client devices has been a major barrier when we wish to train large models with potentially hundreds of millions of parameters. In this paper, we propose a new architecture, referred to as Infocomm, that incorporates locally supervised learning in federated learning. With locally supervised learning, the disadvantages of split learning can be avoided by using a more flexible way to offload training from resource constrained clients to a more capable server. Infocomm enables parallel training of different modules of the neural network in both the server and clients in a gradient-isolated fashion. The efficacy in reducing both training time and communication time is supported by our theoretical analysis and empirical results. In the scenario involving larger models and fewer available local data, Infocomm has been observed to reduce the elapsed time per round by over 37% without sacrificing accuracy compared to both conventional federated learning or directly combining federated learning and split learning, which showcases the advantages of Infocomm under power-constrained IoT scenarios. IEEE
Recently,intelligent reflecting surfaces(IRSs)have emerged as potential candidates for overcoming the line-of-sight issue in 5 G/6 G wireless *** IRSs can manipulate the direction of reflected beams,enabling efficient...
详细信息
Recently,intelligent reflecting surfaces(IRSs)have emerged as potential candidates for overcoming the line-of-sight issue in 5 G/6 G wireless *** IRSs can manipulate the direction of reflected beams,enabling efficient beam steering to enhance the performance of wireless *** unit cell(or unit structure)of an IRS commonly consists of electrical elements for phase ***,by employing phase modulation alone,an IRS can steer the reflected electromagnetic waves toward only discrete and specific angles,leaving a wide range of out-of-beam *** this work,an IRS that uses both phase modulation and space modulation is presented to improve the beam resolution and continuously cover out-of-beam areas that phase modulation alone cannot address.A positive-intrinsic-negative diode is mounted on a unit cell for phase modulation,and a 4D-printed reconfigured structure is fabricated to demonstrate space *** beam-steering function is achieved by alternating the states of the diodes in the same columns,while the beam resolution is improved by controlling the gaps between the *** functions are frst theoretically and numerically analyzed and then experimentally verified,demonstrating that additional angles of-46°/+50°,-22°/+14°,and -16°/+12°are achieved with space modulation and -60°/+62°,-30°/+22°,and±16°are achieved by phase modulation *** proposed IRS offers the possibility of functional integration in a variety of indoor applications within the wireless communication field.
In the Internet of Things (IoT), optimizing machine performance through data analysis and improved connectivity is pivotal. Addressing the growing need for environmentally friendly IoT solutions, we focus on "gre...
详细信息
This study addresses the critical need for robust safeguarding of sensitive data stored on personal computing devices and during data transmissions, alongside the increasing need for secure digital interactions. Conve...
详细信息
MXene is a promising energy storage material for miniaturized microbatteries and microsupercapacitors(MSCs).Despite its superior electrochemical performance,only a few studies have reported MXene-based ultrahigh-rate(...
详细信息
MXene is a promising energy storage material for miniaturized microbatteries and microsupercapacitors(MSCs).Despite its superior electrochemical performance,only a few studies have reported MXene-based ultrahigh-rate(>1000 mV s^(−1))on-paper MSCs,mainly due to the reduced electrical conductance of MXene films deposited on ***,ultrahigh-rate metal-free on-paper MSCs based on heterogeneous MXene/poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate)(PEDOT:PSS)-stack electrodes are fabricated through the combination of direct ink writing and femtosecond laser *** a footprint area of only 20 mm^(2),the on-paper MSCs exhibit excellent high-rate capacitive behavior with an areal capacitance of 5.7 mF cm^(−2)and long cycle life(>95%capacitance retention after 10,000 cycles)at a high scan rate of 1000 mV s^(−1),outperforming most of the present on-paper ***,the heterogeneous MXene/PEDOT:PSS electrodes can interconnect individual MSCs into metal-free on-paper MSC arrays,which can also be simultaneously charged/discharged at 1000 mV s^(−1),showing scalable capacitive *** heterogeneous MXene/PEDOT:PSS stacks are a promising electrode structure for on-paper MSCs to serve as ultrafast miniaturized energy storage components for emerging paper electronics.
The rapid growth of digital data necessitates advanced natural language processing(NLP)models like BERT(Bidi-rectional Encoder Representations from Transformers),known for its superior performance in text ***,BERT’s ...
详细信息
The rapid growth of digital data necessitates advanced natural language processing(NLP)models like BERT(Bidi-rectional Encoder Representations from Transformers),known for its superior performance in text ***,BERT’s size and computational demands limit its practicality,especially in resource-constrained *** research compresses the BERT base model for Bengali emotion classification through knowledge distillation(KD),pruning,and quantization *** Bengali being the sixth most spoken language globally,NLP research in this area is *** approach addresses this gap by creating an efficient BERT-based model for Bengali *** have explored 20 combinations for KD,quantization,and pruning,resulting in improved speedup,fewer parameters,and reduced memory *** best results demonstrate significant improvements in both speed and *** instance,in the case of mBERT,we achieved a 3.87×speedup and 4×compression ratio with a combination of Distil+Prune+Quant that reduced parameters from 178 to 46 M,while the memory size decreased from 711 to 178 *** results offer scalable solutions for NLP tasks in various languages and advance the field of model compression,making these models suitable for real-world applications in resource-limited environments.
Direct training of Spiking Neural Networks (SNNs) is a challenging task because of their inherent temporality. Added to it, the vanilla Back-propagation based methods are not applicable either, due to the non-differen...
详细信息
Direct training of Spiking Neural Networks (SNNs) is a challenging task because of their inherent temporality. Added to it, the vanilla Back-propagation based methods are not applicable either, due to the non-differentiability of the spikes in SNNs. Surrogate-Derivative based methods with Backpropagation Through Time (BPTT) address these direct training challenges quite well;however, such methods are not neuromorphic-hardware friendly for the On-chip training of SNNs. Recently formalized Three-Factor based Rules (TFR) for direct local-training of SNNs are neuromorphic-hardware friendly;however, they do not effectively leverage the depth of the SNN architectures (we show it empirically here), thus, are limited. In this work, we present an improved version of a conventional three-factor rule, for local learning in SNNs which effectively leverages depth - in the context of learning features hierarchically. Taking inspiration from the Back-propagation algorithm, we theoretically derive our improved, local, three-factor based learning method, named DALTON (Deep LocAl Learning via local WeighTs and SurrOgate-Derivative TraNsfer), which employs weights and surrogate-derivative transfer from the local layers. Along the lines of TFR, our proposed method DALTON is also amenable to the neuromorphic-hardware implementation. Through extensive experiments on static (MNIST, FMNIST, & CIFAR10) and event-based (N-MNIST, DVS128-Gesture, & DVSCIFAR10) datasets, we show that our proposed local-learning method DALTON makes effective use of the depth in Convolutional SNNs, compared to the vanilla TFR implementation. IEEE
This study proposes a real-time integrated framework for LiDAR-based object tracking in autonomous driving environments. Advancements in LiDAR sensors are increasing point cloud data collection, leading to a demand fo...
详细信息
In the realm of deep learning, Generative Adversarial Networks (GANs) have emerged as a topic of significant interest for their potential to enhance model performance and enable effective data augmentation. This paper...
详细信息
Offset quadrature amplitude modulation-based filter bank multicarrier (FBMC-OQAM) method is a promising technology for future wireless communication systems. It offers several advantages over traditional orthogonal fr...
详细信息
暂无评论