arXiv subjects

Jihong Park

Publications and source records attributed to Jihong Park.

At least 127 records · Page 7Linked to original sources

arXiv2020

Harnessing Wireless Channels for Scalable and Privacy-Preserving Federated Learning

Wireless connectivity is instrumental in enabling scalable federated learning (FL), yet wireless channels bring challenges for model training, in which channel randomness perturbs each worker's model update while multiple workers' updates incur significant interference under limited bandwidth. To address these challenges, in this work we formulate a novel constrained optimization problem, and propose an FL framework harnessing wireless channel perturbations and interference for improving privacy, bandwidth-efficiency, and scalability. The resultant algorithm is coined analog federated ADMM (A-FADMM) based on analog transmissions and the alternating direction method of multipliers (ADMM). In A-FADMM, all workers upload their model updates to the parameter server (PS) using a single channel via analog transmissions, during which all models are perturbed and aggregated over-the-air. This not only saves communication bandwidth, but also hides each worker's exact model update trajectory from any eavesdropper including the honest-but-curious PS, thereby preserving data privacy against model inversion attacks. We formally prove the convergence and privacy guarantees of A-FADMM for convex functions under time-varying channels, and numerically show the effectiveness of A-FADMM under noisy channels and stochastic non-convex functions, in terms of convergence speed and scalability, as well as communication bandwidth and energy efficiency.

cs.LG↗

arXiv2020

Federated Knowledge Distillation

Distributed learning frameworks often rely on exchanging model parameters across workers, instead of revealing their raw data. A prime example is federated learning that exchanges the gradients or weights of each neural network model. Under limited communication resources, however, such a method becomes extremely costly particularly for modern deep neural networks having a huge number of model parameters. In this regard, federated distillation (FD) is a compelling distributed learning solution that only exchanges the model outputs whose dimensions are commonly much smaller than the model sizes (e.g., 10 labels in the MNIST dataset). The goal of this chapter is to provide a deep understanding of FD while demonstrating its communication efficiency and applicability to a variety of tasks. To this end, towards demystifying the operational principle of FD, the first part of this chapter provides a novel asymptotic analysis for two foundational algorithms of FD, namely knowledge distillation (KD) and co-distillation (CD), by exploiting the theory of neural tangent kernel (NTK). Next, the second part elaborates on a baseline implementation of FD for a classification task, and illustrates its performance in terms of accuracy and communication efficiency compared to FL. Lastly, to demonstrate the applicability of FD to various distributed learning tasks and environments, the third part presents two selected applications, namely FD over asymmetric uplink-and-downlink wireless channels and FD for reinforcement learning.

cs.LG↗

arXiv2020

Explore-Before-Talk: Multichannel Selection Diversity for Uplink Transmissions in Machine-Type Communication

Improving the data rate of machine-type communication (MTC) is essential in supporting emerging Internet of things (IoT) applications ranging from real-time surveillance to edge machine learning. To this end, in this paper we propose a resource allocation approach for uplink transmissions within a random access procedure in MTC by exploiting multichannel selection diversity, coined explore-before-talk (EBT). Each user in EBT first sends pilot signals through multiple channels that are initially allocated by a base station (BS) for exploration, and then the BS informs a subset of initially allocated channels that are associated with high signal-to-noise ratios (SNRs) for data packet transmission by the user while releasing the rest of the channels for other users. Consequently, EBT exploits a multichannel selection diversity gain during data packet transmission, at the cost of exploration during pilot transmission. We optimize this exploration-exploitation trade-off, by deriving closed-form mean data rate and resource outage probability expressions. Numerical results corroborate that EBT achieves a higher mean data rate while satisfying the same outage constraint, compared to a conventional MTC protocol without exploration.

cs.IT↗

arXiv2020

Integrating LEO Satellites and Multi-UAV Reinforcement Learning for Hybrid FSO/RF Non-Terrestrial Networks

A mega-constellation of low-altitude earth orbit (LEO) satellites (SATs) and burgeoning unmanned aerial vehicles (UAVs) are promising enablers for high-speed and long-distance communications in beyond fifth-generation (5G) systems. Integrating SATs and UAVs within a non-terrestrial network (NTN), in this article we investigate the problem of forwarding packets between two faraway ground terminals through SAT and UAV relays using either millimeter-wave (mmWave) radio-frequency (RF) or free-space optical (FSO) link. Towards maximizing the communication efficiency, the real-time associations with orbiting SATs and the moving trajectories of UAVs should be optimized with suitable FSO/RF links, which is challenging due to the time-varying network topology and a huge number of possible control actions. To overcome the difficulty, we lift this problem to multi-agent deep reinforcement learning (MARL) with a novel action dimensionality reduction technique. Simulation results corroborate that our proposed SAT-UAV integrated scheme achieves 1.99x higher end-to-end sum throughput compared to a benchmark scheme with fixed ground relays. While improving the throughput, our proposed scheme also aims to reduce the UAV control energy, yielding 2.25x higher energy efficiency than a baseline method only maximizing the throughput. Lastly, thanks to utilizing hybrid FSO/RF links, the proposed scheme achieves up to 62.56x higher peak throughput and 21.09x higher worst-case throughput than the cases utilizing either RF or FSO links, highlighting the importance of co-designing SAT-UAV associations, UAV trajectories, and hybrid FSO/RF links in beyond-5G NTNs.

cs.NI↗

arXiv2020

When Wireless Communications Meet Computer Vision in Beyond 5G

This article articulates the emerging paradigm, sitting at the confluence of computer vision and wireless communication, to enable beyond-5G/6G mission-critical applications (autonomous/remote-controlled vehicles, visuo-haptic VR, and other cyber-physical applications). First, drawing on recent advances in machine learning and the availability of non-RF data, vision-aided wireless networks are shown to significantly enhance the reliability of wireless communication without sacrificing spectral efficiency. In particular, we demonstrate how computer vision enables {look-ahead} prediction in a millimeter-wave channel blockage scenario, before the blockage actually happens. From a computer vision perspective, we highlight how radio frequency (RF) based sensing and imaging are instrumental in robustifying computer vision applications against occlusion and failure. This is corroborated via an RF-based image reconstruction use case, showcasing a receiver-side image failure correction resulting in reduced retransmission and latency. Taken together, this article sheds light on the much-needed convergence of RF and non-RF modalities to enable ultra-reliable communication and truly intelligent 6G networks.

cs.CV↗

arXiv2020

Communication-Efficient Massive UAV Online Path Control: Federated Learning Meets Mean-Field Game Theory

This paper investigates the control of a massive population of UAVs such as drones. The straightforward method of control of UAVs by considering the interactions among them to make a flock requires a huge inter-UAV communication which is impossible to implement in real-time applications. One method of control is to apply the mean-field game (MFG) framework which substantially reduces communications among the UAVs. However, to realize this framework, powerful processors are required to obtain the control laws at different UAVs. This requirement limits the usage of the MFG framework for real-time applications such as massive UAV control. Thus, a function approximator based on neural networks (NN) is utilized to approximate the solutions of Hamilton-Jacobi-Bellman (HJB) and Fokker-Planck-Kolmogorov (FPK) equations. Nevertheless, using an approximate solution can violate the conditions for convergence of the MFG framework. Therefore, the federated learning (FL) approach which can share the model parameters of NNs at drones, is proposed with NN based MFG to satisfy the required conditions. The stability analysis of the NN based MFG approach is presented and the performance of the proposed FL-MFG is elaborated by the simulations.

cs.NI↗

arXiv2020

Communication-Efficient and Distributed Learning Over Wireless Networks: Principles and Applications

Machine learning (ML) is a promising enabler for the fifth generation (5G) communication systems and beyond. By imbuing intelligence into the network edge, edge nodes can proactively carry out decision-making, and thereby react to local environmental changes and disturbances while experiencing zero communication latency. To achieve this goal, it is essential to cater for high ML inference accuracy at scale under time-varying channel and network dynamics, by continuously exchanging fresh data and ML model updates in a distributed way. Taming this new kind of data traffic boils down to improving the communication efficiency of distributed learning by optimizing communication payload types, transmission techniques, and scheduling, as well as ML architectures, algorithms, and data processing methods. To this end, this article aims to provide a holistic overview of relevant communication and ML principles, and thereby present communication-efficient and distributed learning frameworks with selected use cases.

cs.LG↗

arXiv2020

Towards Enabling Critical mMTC: A Review of URLLC within mMTC

Massive machine-type communication (mMTC) and ultra-reliable and low-latency communication (URLLC) are two key service types in the fifth-generation (5G) communication systems, pursuing scalability and reliability with low-latency, respectively. These two extreme services are envisaged to agglomerate together into \emph{critical mMTC} shortly with emerging use cases (e.g., wide-area disaster monitoring, wireless factory automation), creating new challenges to designing wireless systems beyond 5G. While conventional network slicing is effective in supporting a simple mixture of mMTC and URLLC, it is difficult to simultaneously guarantee the reliability, latency, and scalability requirements of critical mMTC (e.g., < 4ms latency, $10^6$ devices/km$^2$ for factory automation) with limited radio resources. Furthermore, recently proposed solutions to scalable URLLC (e.g., machine learning aided URLLC for driverless vehicles) are ill-suited to critical mMTC whose machine type users have minimal energy budget and computing capability that should be (tightly) optimized for given tasks. To this end, our paper aims to characterize promising use cases of critical mMTC and search for their possible solutions. To this end, we first review the state-of-the-art (SOTA) technologies for separate mMTC and URLLC services and then identify key challenges from conflicting SOTA requirements, followed by potential approaches to prospective critical mMTC solutions at different layers.

cs.NI↗

arXiv2020

Distributed Heteromodal Split Learning for Vision Aided mmWave Received Power Prediction

The goal of this work is the accurate prediction of millimeter-wave received power leveraging both radio frequency (RF) signals and heterogeneous visual data from multiple distributed cameras, in a communication and energy-efficient manner while preserving data privacy. To this end, firstly focusing on data privacy, we propose heteromodal split learning with feature aggregation (HetSLAgg) that splits neural network (NN) models into camera-side and base station (BS)-side segments. The BS-side NN segment fuses RF signals and uploaded image features without collecting raw images. However, the usage of multiple visual data leads to an increase in NN input dimensions, which gives rise to additional communication and energy costs. To overcome additional communication and energy costs due to image interpolation to blend different frame rates, we propose a novel BS-side manifold mixup technique that offloads the interpolation operations from cameras to a BS. Subsequently, we confront energy costs for operating a larger size of the BS- side NN segment due to concatenating image features across cameras and propose an energy-efficient aggregation method. This is done via a linear combination of image features instead of concatenating them, where the NN size is independent of the number of cameras. Comprehensive test-bed experiments with measured channels demonstrate that HetSLAgg reduces the prediction error by 44% compared to a baseline leveraging only RF received power. Moreover, the experiments show that the designed HetSLAgg achieves over 20% gains in terms of communication and energy cost reduction compared to several baseline designs within at most 1% of accuracy loss.

cs.NI↗

arXiv2020

L-FGADMM: Layer-Wise Federated Group ADMM for Communication Efficient Decentralized Deep Learning

This article proposes a communication-efficient decentralized deep learning algorithm, coined layer-wise federated group ADMM (L-FGADMM). To minimize an empirical risk, every worker in L-FGADMM periodically communicates with two neighbors, in which the periods are separately adjusted for different layers of its deep neural network. A constrained optimization problem for this setting is formulated and solved using the stochastic version of GADMM proposed in our prior work. Numerical evaluations show that by less frequently exchanging the largest layer, L-FGADMM can significantly reduce the communication cost, without compromising the convergence speed. Surprisingly, despite less exchanged information and decentralized operations, intermittently skipping the largest layer consensus in L-FGADMM creates a regularizing effect, thereby achieving the test accuracy as high as federated learning (FL), a baseline method with the entire layer consensus by the aid of a central entity.

cs.LG↗

arXiv2020

Mix2FLD: Downlink Federated Learning After Uplink Federated Distillation With Two-Way Mixup

This letter proposes a novel communication-efficient and privacy-preserving distributed machine learning framework, coined Mix2FLD. To address uplink-downlink capacity asymmetry, local model outputs are uploaded to a server in the uplink as in federated distillation (FD), whereas global model parameters are downloaded in the downlink as in federated learning (FL). This requires a model output-to-parameter conversion at the server, after collecting additional data samples from devices. To preserve privacy while not compromising accuracy, linearly mixed-up local samples are uploaded, and inversely mixed up across different devices at the server. Numerical evaluations show that Mix2FLD achieves up to 16.7% higher test accuracy while reducing convergence time by up to 18.8% under asymmetric uplink-downlink channels compared to FL.

cs.LG↗

arXiv2020

XOR Mixup: Privacy-Preserving Data Augmentation for One-Shot Federated Learning

User-generated data distributions are often imbalanced across devices and labels, hampering the performance of federated learning (FL). To remedy to this non-independent and identically distributed (non-IID) data problem, in this work we develop a privacy-preserving XOR based mixup data augmentation technique, coined XorMixup, and thereby propose a novel one-shot FL framework, termed XorMixFL. The core idea is to collect other devices' encoded data samples that are decoded only using each device's own data samples. The decoding provides synthetic-but-realistic samples until inducing an IID dataset, used for model training. Both encoding and decoding procedures follow the bit-wise XOR operations that intentionally distort raw samples, thereby preserving data privacy. Simulation results corroborate that XorMixFL achieves up to 17.6% higher accuracy than Vanilla FL under a non-IID MNIST dataset.

cs.LG↗

arXiv2020

Integrating LEO Satellite and UAV Relaying via Reinforcement Learning for Non-Terrestrial Networks

A mega-constellation of low-earth orbit (LEO) satellites has the potential to enable long-range communication with low latency. Integrating this with burgeoning unmanned aerial vehicle (UAV) assisted non-terrestrial networks will be a disruptive solution for beyond 5G systems provisioning large scale three-dimensional connectivity. In this article, we study the problem of forwarding packets between two faraway ground terminals, through an LEO satellite selected from an orbiting constellation and a mobile high-altitude platform (HAP) such as a fixed-wing UAV. To maximize the end-to-end data rate, the satellite association and HAP location should be optimized, which is challenging due to a huge number of orbiting satellites and the resulting time-varying network topology. We tackle this problem using deep reinforcement learning (DRL) with a novel action dimension reduction technique. Simulation results corroborate that our proposed method achieves up to 5.74x higher average data rate compared to a direct communication baseline without SAT and HAP.

cs.NI↗

arXiv2020

Communication and Consensus Co-Design for Distributed, Low-Latency and Reliable Wireless Systems

Designing distributed, fast and reliable wireless consensus protocols is instrumental in enabling mission-critical decentralized systems, such as robotic networks in the industrial Internet of Things (IIoT), drone swarms in rescue missions, and so forth. However, chasing both low-latency and reliability of consensus protocols is a challenging task. The problem is aggravated under wireless connectivity that may be slower and less reliable, compared to wired connections. To tackle this issue, we investigate fundamental relationships between consensus latency and reliability through the lens of wireless connectivity, and co-design communication and consensus protocols for low-latency and reliable decentralized systems. Specifically, we propose a novel communication-efficient distributed consensus protocol, termed Random Representative Consensus (R2C), and show its effectiveness under gossip and broadcast communication protocols. To this end, we derive a closed-form end-to-end (E2E) latency expression of the R2C that guarantees a target reliability, and compare it with a baseline consensus protocol, referred to as Referendum Consensus (RC). The result shows that the R2C is faster compared to the RC and more reliable compared when co-designed with the broadcast protocol compared to that with the gossip protocol.

cs.IT↗

arXiv2020

Proxy Experience Replay: Federated Distillation for Distributed Reinforcement Learning

Traditional distributed deep reinforcement learning (RL) commonly relies on exchanging the experience replay memory (RM) of each agent. Since the RM contains all state observations and action policy history, it may incur huge communication overhead while violating the privacy of each agent. Alternatively, this article presents a communication-efficient and privacy-preserving distributed RL framework, coined federated reinforcement distillation (FRD). In FRD, each agent exchanges its proxy experience replay memory (ProxRM), in which policies are locally averaged with respect to proxy states clustering actual states. To provide FRD design insights, we present ablation studies on the impact of ProxRM structures, neural network architectures, and communication intervals. Furthermore, we propose an improved version of FRD, coined mixup augmented FRD (MixFRD), in which ProxRM is interpolated using the mixup data augmentation algorithm. Simulations in a Cartpole environment validate the effectiveness of MixFRD in reducing the variance of mission completion time and communication cost, compared to the benchmark schemes, vanilla FRD, federated reinforcement learning (FRL), and policy distillation (PD).

cs.LG↗

arXiv2020

Federated Reinforcement Distillation with Proxy Experience Memory

In distributed reinforcement learning, it is common to exchange the experience memory of each agent and thereby collectively train their local models. The experience memory, however, contains all the preceding state observations and their corresponding policies of the host agent, which may violate the privacy of the agent. To avoid this problem, in this work, we propose a privacy-preserving distributed reinforcement learning (RL) framework, termed federated reinforcement distillation (FRD). The key idea is to exchange a proxy experience memory comprising a pre-arranged set of states and time-averaged policies, thereby preserving the privacy of actual experiences. Based on an advantage actor-critic RL architecture, we numerically evaluate the effectiveness of FRD and investigate how the performance of FRD is affected by the proxy memory structure and different memory exchanging rules.

cs.LG↗

arXiv2020

GADMM: Fast and Communication Efficient Framework for Distributed Machine Learning

When the data is distributed across multiple servers, lowering the communication cost between the servers (or workers) while solving the distributed learning problem is an important problem and is the focus of this paper. In particular, we propose a fast, and communication-efficient decentralized framework to solve the distributed machine learning (DML) problem. The proposed algorithm, Group Alternating Direction Method of Multipliers (GADMM) is based on the Alternating Direction Method of Multipliers (ADMM) framework. The key novelty in GADMM is that it solves the problem in a decentralized topology where at most half of the workers are competing for the limited communication resources at any given time. Moreover, each worker exchanges the locally trained model only with two neighboring workers, thereby training a global model with a lower amount of communication overhead in each exchange. We prove that GADMM converges to the optimal solution for convex loss functions, and numerically show that it converges faster and more communication-efficient than the state-of-the-art communication-efficient algorithms such as the Lazily Aggregated Gradient (LAG) and dual averaging, in linear and logistic regression tasks on synthetic and real datasets. Furthermore, we propose Dynamic GADMM (D-GADMM), a variant of GADMM, and prove its convergence under the time-varying network topology of the workers.

cs.LG↗

arXiv2020

Communication-Efficient Multimodal Split Learning for mmWave Received Power Prediction

The goal of this study is to improve the accuracy of millimeter wave received power prediction by utilizing camera images and radio frequency (RF) signals, while gathering image inputs in a communication-efficient and privacy-preserving manner. To this end, we propose a distributed multimodal machine learning (ML) framework, coined multimodal split learning (MultSL), in which a large neural network (NN) is split into two wirelessly connected segments. The upper segment combines images and received powers for future received power prediction, whereas the lower segment extracts features from camera images and compresses its output to reduce communication costs and privacy leakage. Experimental evaluation corroborates that MultSL achieves higher accuracy than the baselines utilizing either images or RF signals. Remarkably, without compromising accuracy, compressing the lower segment output by 16x yields 16x lower communication latency and 2.8% less privacy leakage compared to the case without compression.

cs.NI↗