arXiv subjects

Yuxuan Sun

Publications and source records attributed to Yuxuan Sun.

At least 109 records · Page 6Linked to original sources

arXiv2022

NTIRE 2021 Multi-modal Aerial View Object Classification Challenge

In this paper, we introduce the first Challenge on Multi-modal Aerial View Object Classification (MAVOC) in conjunction with the NTIRE 2021 workshop at CVPR. This challenge is composed of two different tracks using EO andSAR imagery. Both EO and SAR sensors possess different advantages and drawbacks. The purpose of this competition is to analyze how to use both sets of sensory information in complementary ways. We discuss the top methods submitted for this competition and evaluate their results on our blind test set. Our challenge results show significant improvement of more than 15% accuracy from our current baselines for each track of the competition

cs.CV↗

arXiv2022

Time-Correlated Sparsification for Efficient Over-the-Air Model Aggregation in Wireless Federated Learning

Federated edge learning (FEEL) is a promising distributed machine learning (ML) framework to drive edge intelligence applications. However, due to the dynamic wireless environments and the resource limitations of edge devices, communication becomes a major bottleneck. In this work, we propose time-correlated sparsification with hybrid aggregation (TCS-H) for communication-efficient FEEL, which exploits jointly the power of model compression and over-the-air computation. By exploiting the temporal correlations among model parameters, we construct a global sparsification mask, which is identical across devices, and thus enables efficient model aggregation over-the-air. Each device further constructs a local sparse vector to explore its own important parameters, which are aggregated via digital communication with orthogonal multiple access. We further design device scheduling and power allocation algorithms for TCS-H. Experiment results show that, under limited communication resources, TCS-H can achieve significantly higher accuracy compared to the conventional top-K sparsification with orthogonal model aggregation, with both i.i.d. and non-i.i.d. data distributions.

cs.IT↗

arXiv2022

Online V2X Scheduling for Raw-Level Cooperative Perception

Cooperative perception of connected vehicles comes to the rescue when the field of view restricts stand-alone intelligence. While raw-level cooperative perception preserves most information to guarantee accuracy, it is demanding in communication bandwidth and computation power. Therefore, it is important to schedule the most beneficial vehicle to share its sensor in terms of supplementary view and stable network connection. In this paper, we present a model of raw-level cooperative perception and formulate the energy minimization problem of sensor sharing scheduling as a variant of the Multi-Armed Bandit (MAB) problem. Specifically, volatility of the neighboring vehicles, heterogeneity of V2X channels, and the time-varying traffic context are taken into consideration. Then we propose an online learning-based algorithm with logarithmic performance loss, achieving a decent trade-off between exploration and exploitation. Simulation results under different scenarios indicate that the proposed algorithm quickly learns to schedule the optimal cooperative vehicle and saves more energy as compared to baseline algorithms.

cs.IT↗

arXiv2021

NeurIPS 2021 Competition IGLU: Interactive Grounded Language Understanding in a Collaborative Environment

Human intelligence has the remarkable ability to adapt to new tasks and environments quickly. Starting from a very young age, humans acquire new skills and learn how to solve new tasks either by imitating the behavior of others or by following provided natural language instructions. To facilitate research in this direction, we propose IGLU: Interactive Grounded Language Understanding in a Collaborative Environment. The primary goal of the competition is to approach the problem of how to build interactive agents that learn to solve a task while provided with grounded natural language instructions in a collaborative environment. Understanding the complexity of the challenge, we split it into sub-tasks to make it feasible for participants. This research challenge is naturally related, but not limited, to two fields of study that are highly relevant to the NeurIPS community: Natural Language Understanding and Generation (NLU/G) and Reinforcement Learning (RL). Therefore, the suggested challenge can bring two communities together to approach one of the important challenges in AI. Another important aspect of the challenge is the dedication to perform a human-in-the-loop evaluation as a final evaluation for the agents developed by contestants.

cs.AI↗

arXiv2021

Coded Computation across Shared Heterogeneous Workers with Communication Delay

Distributed computing enables large-scale computation tasks to be processed over multiple workers in parallel. However, the randomness of communication and computation delays across workers causes the straggler effect, which may degrade the performance. Coded computation helps to mitigate the straggler effect, but the amount of redundant load and their assignment to the workers should be carefully optimized. In this work, we consider a multi-master heterogeneous-worker distributed computing scenario, where multiple matrix multiplication tasks are encoded and allocated to workers for parallel computation. The goal is to minimize the communication plus computation delay of the slowest task. We propose worker assignment, resource allocation and load allocation algorithms under both dedicated and fractional worker assignment policies, where each worker can process the encoded tasks of either a single master or multiple masters, respectively. Then, the non-convex delay minimization problem is solved by employing the Markov's inequality-based approximation, Karush-Kuhn-Tucker conditions, and successive convex approximation methods. Through extensive simulations, we show that the proposed algorithms can reduce the task completion delay compared to the benchmarks, and observe that dedicated and fractional worker assignment policies have different scopes of applications.

cs.DC↗

arXiv2021

Dynamic Scheduling for Over-the-Air Federated Edge Learning with Energy Constraints

Machine learning and wireless communication technologies are jointly facilitating an intelligent edge, where federated edge learning (FEEL) is a promising training framework. As wireless devices involved in FEEL are resource limited in terms of communication bandwidth, computing power and battery capacity, it is important to carefully schedule them to optimize the training performance. In this work, we consider an over-the-air FEEL system with analog gradient aggregation, and propose an energy-aware dynamic device scheduling algorithm to optimize the training performance under energy constraints of devices, where both communication energy for gradient aggregation and computation energy for local training are included. The consideration of computation energy makes dynamic scheduling challenging, as devices are scheduled before local training, but the communication energy for over-the-air aggregation depends on the l2-norm of local gradient, which is known after local training. We thus incorporate estimation methods into scheduling to predict the gradient norm. Taking the estimation error into account, we characterize the performance gap between the proposed algorithm and its offline counterpart. Experimental results show that, under a highly unbalanced local data distribution, the proposed algorithm can increase the accuracy by 4.9% on CIFAR-10 dataset compared with the myopic benchmark, while satisfying the energy constraints.

cs.LG↗

arXiv2021

droidlet: modular, heterogenous, multi-modal agents

In recent years, there have been significant advances in building end-to-end Machine Learning (ML) systems that learn at scale. But most of these systems are: (a) isolated (perception, speech, or language only); (b) trained on static datasets. On the other hand, in the field of robotics, large-scale learning has always been difficult. Supervision is hard to gather and real world physical interactions are expensive. In this work we introduce and open-source droidlet, a modular, heterogeneous agent architecture and platform. It allows us to exploit both large-scale static datasets in perception and language and sophisticated heuristics often used in robotics; and provides tools for interactive annotation. Furthermore, it brings together perception, language and action onto one platform, providing a path towards agents that learn from the richness of real world interactions.

cs.RO↗

arXiv2020

On curves with high multiplicity on $\mathbb{P}(a,b,c)$ for $\min(a,b,c)\leq4$

On a weighted projective surface $\mathbb{P}(a,b,c)$ with $\min(a,b,c)\leq 4$, we compute lower bounds for the {\em effective threshold} of an ample divisor, in other words, the highest multiplicity a section of the divisor can have at a specified point. We expect that these bounds are close to being sharp. This translates into finding divisor classes on the blowup of $\mathbb{P}(a,b,c)$ that generate a cone contained in, and probably close to, the effective cone.

math.AG↗

arXiv2020

Edge Learning with Timeliness Constraints: Challenges and Solutions

Future machine learning (ML) powered applications, such as autonomous driving and augmented reality, involve training and inference tasks with timeliness requirements and are communication and computation intensive, which demands for the edge learning framework. The real-time requirements drive us to go beyond accuracy for ML. In this article, we introduce the concept of timely edge learning, aiming to achieve accurate training and inference while minimizing the communication and computation delay. We discuss key challenges and propose corresponding solutions from data, model and resource management perspectives to meet the timeliness requirements. Particularly, for edge training, we argue that the total training delay rather than rounds should be considered, and propose data or model compression, and joint device scheduling and resource management schemes for both centralized training and federated learning systems. For edge inference, we explore the dependency between accuracy and delay for communication and computation, and propose dynamic data compression and flexible pruning schemes. Two case studies show that the timeliness performances, including the training accuracy under a given delay budget and the completion ratio of inference tasks within deadline, are highly improved with the proposed solutions.

cs.NI↗

arXiv2020

Distributed Task Replication for Vehicular Edge Computing: Performance Analysis and Learning-based Algorithm

In a vehicular edge computing (VEC) system, vehicles can share their surplus computation resources to provide cloud computing services. The highly dynamic environment of the vehicular network makes it challenging to guarantee the task offloading delay. To this end, we introduce task replication to the VEC system, where the replicas of a task are offloaded to multiple vehicles at the same time, and the task is completed upon the first response among replicas. First, the impact of the number of task replicas on the offloading delay is characterized, and the optimal number of task replicas is approximated in closed-form. Based on the analytical result, we design a learning-based task replication algorithm (LTRA) with combinatorial multi-armed bandit theory, which works in a distributed manner and can automatically adapt itself to the dynamics of the VEC system. A realistic traffic scenario is used to evaluate the delay performance of the proposed algorithm. Results show that, under our simulation settings, LTRA with an optimized number of task replicas can reduce the average offloading delay by over 30% compared to the benchmark without task replication, and at the same time can improve the task completion ratio from 97% to 99.6%.

cs.NI↗

arXiv2019

ROI Pooled Correlation Filters for Visual Tracking

The ROI (region-of-interest) based pooling method performs pooling operations on the cropped ROI regions for various samples and has shown great success in the object detection methods. It compresses the model size while preserving the localization accuracy, thus it is useful in the visual tracking field. Though being effective, the ROI-based pooling operation is not yet considered in the correlation filter formula. In this paper, we propose a novel ROI pooled correlation filter (RPCF) algorithm for robust visual tracking. Through mathematical derivations, we show that the ROI-based pooling can be equivalently achieved by enforcing additional constraints on the learned filter weights, which makes the ROI-based pooling feasible on the virtual circular samples. Besides, we develop an efficient joint training formula for the proposed correlation filter algorithm, and derive the Fourier solvers for efficient model training. Finally, we evaluate our RPCF tracker on OTB-2013, OTB-2015 and VOT-2017 benchmark datasets. Experimental results show that our tracker performs favourably against other state-of-the-art trackers.

cs.CV↗

arXiv2019

Energy-Aware Analog Aggregation for Federated Learning with Redundant Data

Federated learning (FL) enables workers to learn a model collaboratively by using their local data, with the help of a parameter server (PS) for global model aggregation. The high communication cost for periodic model updates and the non-independent and identically distributed (i.i.d.) data become major bottlenecks for FL. In this work, we consider analog aggregation to scale down the communication cost with respect to the number of workers, and introduce data redundancy to the system to deal with non-i.i.d. data. We propose an online energy-aware dynamic worker scheduling policy, which maximizes the average number of workers scheduled for gradient update at each iteration under a long-term energy constraint, and analyze its performance based on Lyapunov optimization. Experiments using MNIST dataset show that, for non-i.i.d. data, doubling data storage can improve the accuracy by 9.8% under a stringent energy budget, while the proposed policy can achieve close-to-optimal accuracy without violating the energy constraint.

cs.IT↗

arXiv2019

Heterogeneous Coded Computation across Heterogeneous Workers

Coded distributed computing framework enables large-scale machine learning (ML) models to be trained efficiently in a distributed manner, while mitigating the straggler effect. In this work, we consider a multi-task assignment problem in a coded distributed computing system, where multiple masters, each with a different matrix multiplication task, assign computation tasks to workers with heterogeneous computing capabilities. Both dedicated and probabilistic worker assignment models are considered, with the objective of minimizing the average completion time of all computations. For dedicated worker assignment, greedy algorithms are proposed and the corresponding optimal load allocation is derived based on the Lagrange multiplier method. For probabilistic assignment, successive convex approximation method is used to solve the non-convex optimization problem. Simulation results show that the proposed algorithms reduce the completion time by 80% over uncoded scheme, and 49% over an unbalanced coded scheme.

cs.IT↗

arXiv2019

Exploiting Moving Intelligence: Delay-Optimized Computation Offloading in Vehicular Fog Networks

Future vehicles will have rich computing resources to support autonomous driving and be connected by wireless technologies. Vehicular fog networks (VeFN) have thus emerged to enable computing resource sharing via computation task offloading, providing wide range of fog applications. However, the high mobility of vehicles makes it hard to guarantee the delay that accounts for both communication and computation throughout the whole task offloading procedure. In this article, we first review the state-of-the-art of task offloading in VeFN, and argue that mobility is not only an obstacle for timely computing in VeFN, but can also benefit the delay performance. We then identify machine learning and coded computing as key enabling technologies to address and exploit mobility in VeFN. Case studies are provided to illustrate how to adapt learning algorithms to fit for the dynamic environment in VeFN, and how to exploit the mobility with opportunistic computation offloading and task replication.

cs.NI↗

arXiv2019

Adaptive Learning-Based Task Offloading for Vehicular Edge Computing Systems

The vehicular edge computing (VEC) system integrates the computing resources of vehicles, and provides computing services for other vehicles and pedestrians with task offloading. However, the vehicular task offloading environment is dynamic and uncertain, with fast varying network topologies, wireless channel states and computing workloads. These uncertainties bring extra challenges to task offloading. In this work, we consider the task offloading among vehicles, and propose a solution that enables vehicles to learn the offloading delay performance of their neighboring vehicles while offloading computation tasks. We design an adaptive learning-based task offloading (ALTO) algorithm based on the multi-armed bandit (MAB) theory, in order to minimize the average offloading delay. ALTO works in a distributed manner without requiring frequent state exchange, and is augmented with input-awareness and occurrence-awareness to adapt to the dynamic environment. The proposed algorithm is proved to have a sublinear learning regret. Extensive simulations are carried out under both synthetic scenario and realistic highway scenario, and results illustrate that the proposed algorithm achieves low delay performance, and decreases the average delay up to 30% compared with the existing upper confidence bound based learning algorithm.

cs.IT↗

arXiv2018

Task Replication for Vehicular Edge Computing: A Combinatorial Multi-Armed Bandit based Approach

In vehicular edge computing (VEC) system, some vehicles with surplus computing resources can provide computation task offloading opportunities for other vehicles or pedestrians. However, vehicular network is highly dynamic, with fast varying channel states and computation loads. These dynamics are difficult to model or to predict, but they have major impact on the quality of service (QoS) of task offloading, including delay performance and service reliability. Meanwhile, the computing resources in VEC are often redundant due to the high density of vehicles. To improve the QoS of VEC and exploit the abundant computing resources on vehicles, we propose a learning-based task replication algorithm (LTRA) based on combinatorial multi-armed bandit (CMAB) theory, in order to minimize the average offloading delay. LTRA enables multiple vehicles to process the replicas of the same task simultaneously, and vehicles that require computing services can learn the delay performance of other vehicles while offloading tasks. We take the occurrence time of vehicles into consideration, and redesign the utility function of existing CMAB algorithm, so that LTRA can adapt to the time varying network topology of VEC. We use a realistic highway scenario to evaluate the delay performance and service reliability of LTRA through simulations, and show that compared with single task offloading, LTRA can improve the task completion ratio with deadline 0.6s from 80% to 98%.

cs.IT↗

arXiv2018

Learning-Based Task Offloading for Vehicular Cloud Computing Systems

Vehicular cloud computing (VCC) is proposed to effectively utilize and share the computing and storage resources on vehicles. However, due to the mobility of vehicles, the network topology, the wireless channel states and the available computing resources vary rapidly and are difficult to predict. In this work, we develop a learning-based task offloading framework using the multi-armed bandit (MAB) theory, which enables vehicles to learn the potential task offloading performance of its neighboring vehicles with excessive computing resources, namely service vehicles (SeVs), and minimizes the average offloading delay. We propose an adaptive volatile upper confidence bound (AVUCB) algorithm and augment it with load-awareness and occurrence-awareness, by redesigning the utility function of the classic MAB algorithms. The proposed AVUCB algorithm can effectively adapt to the dynamic vehicular environment, balance the tradeoff between exploration and exploitation in the learning process, and converge fast to the optimal SeV with theoretical performance guarantee. Simulations under both synthetic scenario and a realistic highway scenario are carried out, showing that the proposed algorithm achieves close-to-optimal delay performance.

cs.IT↗

arXiv2017

EMM: Energy-Aware Mobility Management for Mobile Edge Computing in Ultra Dense Networks

Merging mobile edge computing (MEC) functionality with the dense deployment of base stations (BSs) provides enormous benefits such as a real proximity, low latency access to computing resources. However, the envisioned integration creates many new challenges, among which mobility management (MM) is a critical one. Simply applying existing radio access oriented MM schemes leads to poor performance mainly due to the co-provisioning of radio access and computing services of the MEC-enabled BSs. In this paper, we develop a novel user-centric energy-aware mobility management (EMM) scheme, in order to optimize the delay due to both radio access and computation, under the long-term energy consumption constraint of the user. Based on Lyapunov optimization and multi-armed bandit theories, EMM works in an online fashion without future system state information, and effectively handles the imperfect system state information. Theoretical analysis explicitly takes radio handover and computation migration cost into consideration and proves a bounded deviation on both the delay performance and energy consumption compared to the oracle solution with exact and complete future system information. The proposed algorithm also effectively handles the scenario in which candidate BSs randomly switch on/off during the offloading process of a task. Simulations show that the proposed algorithms can achieve close-to-optimal delay performance while satisfying the user energy consumption constraint.

cs.IT↗