https://sites.temple.edu/pdames/files/2016/07/DamesTokekarKumarISRR2015.pdf
Abstract Target tracking is a fundamental problem in robotics research and has been the subject of detailed studies over the years. In this paper, we introduce a new formulation, based on the mathematical concept of random finite sets, that allows for tracking an unknown and dynamic number of mobile targets with a team of robots. We show how to employ the Probability Hypothesis Density filter ...
https://cis.temple.edu/~jiewu/research/publications/Publication_files/INFOCOM2024_PSFL%20Parallel-Sequential%20Federated%20Learning%20with%20Convergence%20Guarantees.pdf
Abstract—Federated Learning (FL) is a novel distributed learning paradigm which can coordinate multiple clients to jointly train a machine learning model by using their local data samples. Existing FL works can be roughly divided into two categories according to the modes of model training: Parallel FL (PFL) and Sequential FL (SFL). PFL can speed up each round of model training time through ...
https://cis.temple.edu/~yu/research/PPGSpotter-info24.pdf
Abstract—Free weight training (FWT) is of utmost importance for physical well-being. However, the success of FWT depends heavily on choosing the suitable workload, as improper selections can lead to suboptimal outcomes or injury. Current workload estimation approaches rely on manual recording and special-ized equipment with limited feedback. Therefore, we introduce PPGSpotter, a novel PPG ...
https://cis.temple.edu/~apal/ipic_food.pdf
Computer and Information Sciences, Temple University, Philadelphia, PA 19122 E-mail:famitangshu.pal,kkantg@temple.edu Abstract: In this paper, we introduce a Physical Internet architecture for fresh food distribution networks, with the goal of meeting the key challenges of fresh product delivery and reduce waste. In particular, we explore fuel-efficient delivery of fresh food among different ...
https://cis.temple.edu/~apal/ccgrid_iot.pdf
Abstract—In this paper we examine the problem of conflict detection and mitigation across multiple independently designed IoT subsystems deployed in a shared environment. The desired behavior of the system is codified in terms of predefined “safety properties”. We allow both the operational rules and safety properties to include time and temporal logic operations and detect their ...
https://cis.temple.edu/~wu/research/publications/Publication_files/jsan-13-00044-v2.pdf
Abstract: Harnessing remote computation power over the Internet without the need for expensive hardware and making costly services available to mass users at a marginal cost gave birth to the concept of cloud computing. This survey provides a concise overview of the growing confluence of cloud computing, edge intelligence, and AI, with a focus on their revolutionary impact on the Internet of ...
https://cis.temple.edu/~latecki/Papers/ICIP2019.pdf
ABSTRACT The extensive computational burden limits the usage of CNNs in mobile devices for dense estimation tasks. In this paper, we present a lightweight network to address this prob-lem, namely LEDNet, which employs an asymmetric encoder-decoder architecture for the task of real-time semantic seg-mentation. More specifically, the encoder adopts a ResNet as backbone network, where two new ...
https://cis.temple.edu/~jiewu/research/publications/Publication_files/Privacy-Preserving_Federated_Neural_Architecture_Search_With_Enhanced_Robustness_for_Edge_Computing.pdf
Abstract—With the development of large-scale artificial intelli-gence services, edge devices are becoming essential providers of data and computing power. However, these edge devices are not immune to malicious attacks. Federated learning (FL), while pro-tecting privacy of decentralized data through secure aggregation, struggles to trace adversaries and lacks optimization for hetero-geneity ...
https://cis.temple.edu/~latecki/Papers/Quan_DRBANET_ICIP_2022.pdf
DRBANet follows a dual-branch architecture [12,13], where HRB produces image detail features, while LRB captures im-age semantic cues. DRBANet is built mainly based on EIBM unit, which enables us to explore larger receptive fields, but with very smaller computational overhead.
https://cis.temple.edu/~jiewu/research/publications/Publication_files/m48122-zhao%20final.pdf
ArrayPipe: Introducing Job-Array Pipeline Parallelism for High Throughput Model Exploration Hairui Zhao1, Hongliang Li1,2,∗, Qi Tian1, Jie Wu3, Meng Zhang1, Xiang Li1, Haixiao Xu4