arXiv subjects

Yuqing Du

Publications and source records attributed to Yuqing Du.

27 records · Page 2Linked to original sources

arXiv2020

Robust Reinforcement Learning using Adversarial Populations

Reinforcement Learning (RL) is an effective tool for controller design but can struggle with issues of robustness, failing catastrophically when the underlying system dynamics are perturbed. The Robust RL formulation tackles this by adding worst-case adversarial noise to the dynamics and constructing the noise distribution as the solution to a zero-sum minimax game. However, existing work on learning solutions to the Robust RL formulation has primarily focused on training a single RL agent against a single adversary. In this work, we demonstrate that using a single adversary does not consistently yield robustness to dynamics variations under standard parametrizations of the adversary; the resulting policy is highly exploitable by new adversaries. We propose a population-based augmentation to the Robust RL formulation in which we randomly initialize a population of adversaries and sample from the population uniformly during training. We empirically validate across robotics benchmarks that the use of an adversarial population results in a more robust policy that also improves out-of-distribution generalization. Finally, we demonstrate that this approach provides comparable robustness and generalization as domain randomization on these benchmarks while avoiding a ubiquitous domain randomization failure mode.

cs.LG↗

arXiv2020

Capacity of Remote Classification Over Wireless Channels

Wireless connectivity creates a computing paradigm that merges communication and inference. A basic operation in this paradigm is the one where a device offloads classification tasks to the edge servers. We term this remote classification, with a potential to enable intelligent applications. Remote classification is challenged by the finite and variable data rate of the wireless channel, which affects the capability to transfer high-dimensional features and thus limits the classification resolution. We introduce a set of metrics under the name of classification capacity that are defined as the maximum number of classes that can be discerned over a given communication channel while meeting a target classification error probability. The objective is to choose a subset of classes from a library that offers satisfactory performance over a given channel. We treat two cases of subset selection. First, a device can select the subset by pruning the class library until arriving at a subset that meets the targeted error probability while maximizing the classification capacity. Adopting a subspace data model, we prove the equivalence of classification capacity maximization to Grassmannian packing. The results show that the classification capacity grows exponentially with the instantaneous communication rate, and super-exponentially with the dimensions of each data cluster. This also holds for ergodic and outage capacities with fading if the instantaneous rate is replaced with an average rate and a fixed rate, respectively. In the second case, a device has a preference of class subset for every communication rate, which is modeled as an instance of uniformly sampling the library. Without class selection, the classification capacity and its ergodic and outage counterparts are proved to scale linearly with their corresponding communication rates instead of the exponential growth in the last case.

cs.IT↗

arXiv2020

Energy-Efficient Resource Management for Federated Edge Learning with CPU-GPU Heterogeneous Computing

Edge machine learning involves the deployment of learning algorithms at the network edge to leverage massive distributed data and computation resources to train artificial intelligence (AI) models. Among others, the framework of federated edge learning (FEEL) is popular for its data-privacy preservation. FEEL coordinates global model training at an edge server and local model training at edge devices that are connected by wireless links. This work contributes to the energy-efficient implementation of FEEL in wireless networks by designing joint computation-and-communication resource management ($\text{C}^2$RM). The design targets the state-of-the-art heterogeneous mobile architecture where parallel computing using both a CPU and a GPU, called heterogeneous computing, can significantly improve both the performance and energy efficiency. To minimize the sum energy consumption of devices, we propose a novel $\text{C}^2$RM framework featuring multi-dimensional control including bandwidth allocation, CPU-GPU workload partitioning and speed scaling at each device, and $\text{C}^2$ time division for each link. The key component of the framework is a set of equilibriums in energy rates with respect to different control variables that are proved to exist among devices or between processing units at each device. The results are applied to designing efficient algorithms for computing the optimal $\text{C}^2$RM policies faster than the standard optimization tools. Based on the equilibriums, we further design energy-efficient schemes for device scheduling and greedy spectrum sharing that scavenges "spectrum holes" resulting from heterogeneous $\text{C}^2$ time divisions among devices. Using a real dataset, experiments are conducted to demonstrate the effectiveness of $\text{C}^2$RM on improving the energy efficiency of a FEEL system.

cs.IT↗

arXiv2019

An Introduction to Communication Efficient Edge Machine Learning

In the near future, Internet-of-Things (IoT) is expected to connect billions of devices (e.g., smartphones and sensors), which generate massive real-time data at the network edge. Intelligence can be distilled from the data to support next-generation AI-powered applications, which is called edge machine learning. One challenge faced by edge learning is the communication bottleneck, which is caused by the transmission of high-dimensional data from many edge devices to edge servers for learning. Traditional wireless techniques focusing only on efficient radio access are ineffective in tackling the challenge. Solutions should be based on a new approach that seamlessly integrates communication and computation. This has led to the emergence of a new cross-disciplinary paradigm called communication efficient edge learning. The main theme in the area is to design new communication techniques and protocols for efficient implementation of different distributed learning frameworks (i.e., federated learning) in wireless networks. This article provides an overview of the emerging area by introducing new design principles, discussing promising research opportunities, and providing design examples based on recent work.

cs.IT↗

arXiv2019

High-Dimensional Stochastic Gradient Quantization for Communication-Efficient Edge Learning

Edge machine learning involves the deployment of learning algorithms at the wireless network edge so as to leverage massive mobile data for enabling intelligent applications. The mainstream edge learning approach, federated learning, has been developed based on distributed gradient descent. Based on the approach, stochastic gradients are computed at edge devices and then transmitted to an edge server for updating a global AI model. Since each stochastic gradient is typically high-dimensional (with millions to billions of coefficients), communication overhead becomes a bottleneck for edge learning. To address this issue, we propose in this work a novel framework of hierarchical stochastic gradient quantization and study its effect on the learning performance. First, the framework features a practical hierarchical architecture for decomposing the stochastic gradient into its norm and normalized block gradients, and efficiently quantizes them using a uniform quantizer and a low-dimensional codebook on a Grassmann manifold, respectively. Subsequently, the quantized normalized block gradients are scaled and cascaded to yield the quantized normalized stochastic gradient using a so-called hinge vector designed under the criterion of minimum distortion. The hinge vector is also efficiently compressed using another low-dimensional Grassmannian quantizer. The other feature of the framework is a bit-allocation scheme for reducing the quantization error. The scheme determines the resolutions of the low-dimensional quantizers in the proposed framework. The framework is proved to guarantee model convergency by analyzing the convergence rate as a function of the quantization bits. Furthermore, by simulation, our design is shown to substantially reduce the communication overhead compared with the state-of-the-art signSGD scheme, while both achieve similar learning accuracies.

cs.IT↗

arXiv2019

Energy-Efficient Radio Resource Allocation for Federated Edge Learning

Edge machine learning involves the development of learning algorithms at the network edge to leverage massive distributed data and computation resources. Among others, the framework of federated edge learning (FEEL) is particularly promising for its data-privacy preservation. FEEL coordinates global model training at a server and local model training at edge devices over wireless links. In this work, we explore the new direction of energy-efficient radio resource management (RRM) for FEEL. To reduce devices' energy consumption, we propose energy-efficient strategies for bandwidth allocation and scheduling. They adapt to devices' channel states and computation capacities so as to reduce their sum energy consumption while warranting learning performance. In contrast with the traditional rate-maximization designs, the derived optimal policies allocate more bandwidth to those scheduled devices with weaker channels or poorer computation capacities, which are the bottlenecks of synchronized model updates in FEEL. On the other hand, the scheduling priority function derived in closed form gives preferences to devices with better channels and computation capacities. Substantial energy reduction contributed by the proposed strategies is demonstrated in learning experiments.

cs.IT↗

arXiv2018

Towards an Intelligent Edge: Wireless Communication Meets Machine Learning

The recent revival of artificial intelligence (AI) is revolutionizing almost every branch of science and technology. Given the ubiquitous smart mobile gadgets and Internet of Things (IoT) devices, it is expected that a majority of intelligent applications will be deployed at the edge of wireless networks. This trend has generated strong interests in realizing an "intelligent edge" to support AI-enabled applications at various edge devices. Accordingly, a new research area, called edge learning, emerges, which crosses and revolutionizes two disciplines: wireless communication and machine learning. A major theme in edge learning is to overcome the limited computing power, as well as limited data, at each edge device. This is accomplished by leveraging the mobile edge computing (MEC) platform and exploiting the massive data distributed over a large number of edge devices. In such systems, learning from distributed data and communicating between the edge server and devices are two critical and coupled aspects, and their fusion poses many new research challenges. This article advocates a new set of design principles for wireless communication in edge learning, collectively called learning-driven communication. Illustrative examples are provided to demonstrate the effectiveness of these design principles, and unique research opportunities are identified.

cs.IT↗

arXiv2018

Fast Analog Transmission for High-Mobility Wireless Data Acquisition in Edge Learning

By implementing machine learning at the network edge, edge learning trains models by leveraging rich data distributed at edge devices (e.g., smartphones and sensors) and in return endow on them capabilities of seeing, listening, and reasoning. In edge learning, the need of high-mobility wireless data acquisition arises in scenarios where edge devices (or even servers) are mounted on the ground or aerial vehicles. In this paper, we present a novel solution, called fast analog transmission (FAT), for high- mobility data acquisition in edge-learning systems, which has several key features. First, FAT incurs low-latency. Specifically, FAT requires no source-and-channel coding and no channel training via the proposed technique of Grassmann analog encoding (GAE) that encodes data samples into subspace matrices. Second, FAT supports spatial multiplexing by directly transmitting analog vector data over an antenna array. Third, FAT can be seamlessly integrated with edge learning (i.e., training of a classifier model in this work). In particular, by applying a Grassmannian-classification algorithm from computer vision, the received GAE encoded data can be directly applied to training the model without decoding and conversion. This design is found by simulation to outperform conventional schemes in learning accuracy due to its robustness against data distortion induced by fast fading.

cs.IT↗

arXiv2018

Automatic Recognition of Space-Time Constellations by Learning on the Grassmann Manifold (Extended Version)

Recent breakthroughs in machine learning especially artificial intelligence shift the paradigm of wireless communication towards intelligence radios. One of their core operations is automatic modulation recognition (AMR). Existing research focuses on coherent modulation schemes such as QAM, PSK and FSK. The AMR of (non-coherent) space-time modulation remains an uncharted area despite its wide deployment in modern multiple-input-multiple-output (MIMO) systems. The scheme using a so called Grassmann constellation enables rate-enhancement using multi-antennas and blind detection. In this work, we propose an AMR approach for Grassmann constellation based on data clustering, which differs from traditional AMR based on classification using a modulation database. The approach allows algorithms for clustering on the Grassmann manifold, such as Grassmann K-means and depth-first search, originally developed for computer vision to be applied to AMR. We further develop an analytical framework for studying and designing these algorithms in the context of AMR. First, the maximum-likelihood Grassmann constellation detection is proved to be equivalent to clustering on the Grassmannian. Thereby, a well-known machine-learning result that was originally established only for the Euclidean space is rediscovered for the Grassmannian. Next, despite a rich literature on algorithmic design, theoretical analysis of data clustering is largely overlooked due to the lack of tractable techniques. We tackle the challenge by introducing probabilistic metrics for measuring the inter-cluster separability and intra-cluster connectivity of received space-time symbols and deriving them using tools from differential geometry and Grassmannian packing. The results provide useful insights into the effects of various parameters ranging from the signal-to-noise ratio to constellation size, facilitating algorithmic design.

cs.IT↗