Publications
-
Conference Papers
-
“QoS-Aware Priority-Based Task Offloading for Deep Learning Services at the Edge”
Minoo Hosseinzadeh, Andrew Wachal, Hana Khamfroush, Daniel E. Lucani • IEEE CCNC • 2022
Abstract
Emerging Edge Computing (EC) technology has shown promise for many delay-sensitive Deep Learning (DL) based applications of smart cities in terms of improved Quality-of-Service (QoS). EC requires judicious decisions which jointly consider the limited capacity of the edge servers and provided QoS of DL-dependent services. In a smart city environment, tasks may have varying priorities in terms of when and how to serve them; thus, priorities of the tasks have to be considered when making resource management decisions. In this paper, we focus on finding optimal offloading decisions in a three-tier user-edge-cloud architecture while considering different priority classes for the DL-based services and making a trade-off between a task's completion time and the provided accuracy by the DL-based service. We cast the optimization problem as an Integer Linear Program (ILP) where the objective is to maximize a function called gain of system (GoS) defined based on provided QoS and priority of the tasks. We prove the problem is NP-hard. We then propose an efficient offloading algorithm, called PGUS, that is shown to achieve near-optimal results in terms of the provided GoS. Finally, we compare our proposed algorithm, PGUS, with heuristics and a state-of-the-art algorithm, called GUS, using both numerical analysis and real-world implementation. Our results show that PGUS outperforms GUS by a factor of 45% in average in terms of serving the top 25% higher priority classes of the tasks while still keeping the overall percentage of the dropped tasks minimal and the overall gain of system maximized.
-
“A Framework for Edge Intelligent Smart Distribution Grids via Federated Learning”
Nathaniel Hudson, Md Jakir Hossain, Minoo Hosseinzadeh, Hana Khamfroush, Mahshid Rahnamay-Naeini, Nasir Ghani • IEEE ICCCN • 2021
Abstract
Recent advances in distributed data processing and machine learning provide new opportunities to enable critical, time-sensitive functionalities of smart distribution grids in a secure and reliable fashion. Combining the recent advents of edge computing (EC) and edge intelligence (EI) with existing advanced metering infrastructure (AMI) has the potential to reduce overall communication cost, preserve user privacy, and provide improved situational awareness. In this paper, we provide an overview for how EC and EI can supplement applications relevant to AMI systems. Additionally, using such systems in tandem can enable distributed deep learning frameworks (e.g., federated learning) to empower distributed data processing and intelligent decision making for AMI. Finally, to demonstrate the efficacy of this considered architecture, we approach the non-intrusive load monitoring (NILM) problem using federated learning to train a deep recurrent neural network architecture in a 2-tier and 3-tier manner. In this approach, smart homes locally train a neural network using their metering data and only share the learned model parameters with AMI components for aggregation. Our results show this can reduce communication cost associated with distributed learning, as well as provide an immediate layer of privacy, due to no raw data being communicated to AMI components. Further, we show that FL is able to closely match the model loss provided by standard centralized deep learning where raw data is communicated for centralized training.
-
“QoS-Aware Placement of Deep Learning Services on the Edge with Multiple Service Implementations”
Nathaniel Hudson, Hana Khamfroush, Daniel E. Lucani • IEEE ICCCN Big Data and Machine Learning for Networking (BDMLN) Workshop • 2021
Abstract
Mobile edge computing pushes computationally-intensive services closer to the user to provide reduced delay due to physical proximity. This has led many to consider deploying deep learning models on the edge -- commonly known as edge intelligence (EI). EI services can have many model implementations that provide different QoS. For instance, one model can perform inference faster than another (thus reducing latency) while achieving less accuracy when evaluated. In this paper, we study joint service placement and model scheduling of EI services with the goal to maximize Quality-of-Servcice (QoS) for end users where EI services have multiple implementations to serve user requests, each with varying costs and QoS benefits. We cast the problem as an integer linear program and prove that it is NP-hard. We then prove the objective is equivalent to maximizing a monotone increasing, submodular set function and thus can be solved greedily while maintaining a (1-1/e)-approximation guarantee. We then propose two greedy algorithms: one that theoretically guarantees this approximation and another that empirically matches its performance with greater efficiency. Finally, we thoroughly evaluate the proposed algorithm for making placement and scheduling decisions in both synthetic and real-world scenarios against the optimal solution and some baselines. In the real-world case, we consider real machine learning models using the ImageNet 2012 data-set for requests. Our numerical experiments empirically show that our more efficient greedy algorithm is able to approximate the optimal solution with a 0.904 approximation on average, while the next closest baseline achieves a 0.607 approximation on average.
Citation (BibTex)
@article{hudson2021qos, title={QoS-aware placement of deep learning services on the edge with multiple service implementations}, author={Hudson, Nathaniel and Khamfroush, Hana and Lucani, Daniel E}, journal={arXiv preprint arXiv: 2104.15094}, year={2021}} -
“Optimal Accuracy-Time Trade-off For Deep Learning Services in Edge Computing Systems”
Minoo Hosseinzadeh, Andrew Wachal, Hana Khamfroush, Daniel E. Lucani • IEEE ICC • 2021
Abstract
With the increasing demand for computationally intensive services like deep learning tasks, emerging distributed computing platforms such as edge computing (EC) systems are becoming more popular. Edge computing systems have shown promising results in terms of latency reduction compared to the traditional cloud systems. However, their limited processing capacity imposes a trade-off between the potential latency reduction and the achieved accuracy in computationally-intensive services such as deep learning-based services. In this paper, we focus on finding the optimal accuracy-time trade-off for running deep learning services in a three-tier EC platform where several deep learning models with different accuracy levels are available. Specifically, we cast the problem as an Integer Linear Program, where optimal task scheduling decisions are made to maximize overall user satisfaction in terms of accuracy-time trade-off. We prove that our problem is NP-hard and then provide a polynomial constant-time greedy algorithm, called GUS, that is shown to attain near-optimal results. Finally, upon vetting our algorithmic solution through numerical experiments and comparison with a set of heuristics, we deploy it on a testbed implemented to measure for real-world results. The results of both numerical analysis and real-world implementation show that GUS can outperform the baseline heuristics in terms of the average percentage of satisfied users by a factor of at least 50%.
-
“Improving the Accuracy-Latency Trade-off of Edge-Cloud Computation Offloading for Deep Learning Services”
Xiaobo Zhao, Minoo Hosseinzadeh, Nathaniel Hudson, Hana Khamfroush, Daniel E. Lucani • IEEE GLOBECOM Workshops • 2020
Abstract
Offloading tasks to the edge or the Cloud has the potential to improve accuracy of classification and detection tasks as more powerful hardware and machine learning models can be used. The downside is the added delay introduced for sending the data to the Edge/Cloud. In delay-sensitive applications, it is usually necessary to strike a balance between accuracy and latency. However, the state of the art typically considers offloading all-or-nothing decisions, e.g., process locally or send all available data to the Edge (Cloud). Our goal is to expand the options in the accuracy-latency trade-off by allowing the source to send a fraction of the total data for processing. We evaluate the performance of image classifiers when faced with images that have been purposely reduced in quality in order to reduce traffic costs. Using three common models (SqueezeNet, GoogleNet, ResNet) and two data sets (Caltech101, ImageNet) we show that the Gompertz function provides a good approximation to determine the accuracy of a model given the fraction of the data of the image that is actually conveyed to the model. We formulate the offloading decision process using this new flexibility and show that a better overall accuracy-latency tradeoff is attained: 58% traffic reduction, 25% latency reduction, as well as 12% accuracy improvement.
-
“Edge Layer Design and Optimization for Smart Grids”
Adetola Adeniran, Md Abul Hasnat, Minoo Hosseinzadeh, Hana Khamfroush, Mahshid Rahnamay-Naeini • IEEE SmartGridComm • 2020
Abstract
The emergence of modern monitoring, communication, computation, and control equipment into power systems has made them evolve into smart grids that can be thought of as the electric grid of things. This evolution has enhanced the efficiency of the power systems through the availability of a large volume of system data that can help with system functions nevertheless, it has intensified the communication and computation burden on these systems. While many such computations were traditionally deployed in central servers, new technologies such as edge computing can provide unique opportunities to address some of the computational challenges and improve the responsiveness of the system by processing data locally. In this paper, an edge enabled smart grid architecture is presented. The edge layer for the smart grid is designed through various optimization formulations to identify the placement of edge servers and their connectivity structure to the Phasor Measurement Units in the system. Various factors affecting the design, such as the geographical and resource constraints as well as the communication technology considerations have been incorporated in the formulations and evaluated using the IEEE 118 bus system.
-
“Smart Advertisement for Maximal Clicks in Online Social Networks Without User Data”
Nathaniel Hudson, Hana Khamfroush, Brent Harrison, Adam Craig • IEEE SMARTCOMP • 2020
Abstract
Smart cities are a growing paradigm in the design of systems that interact with one another for informed and efficient decision making, empowered by data and technology, of resources in a city. The diffusion of information to citizens in a smart city will rely on social trends and smart advertisement. Online social networks (OSNs) are prominent and increasingly important platforms to spread information, observe social trends, and advertise new products. To maximize the benefits of such platforms in sharing information, many groups invest in finding ways to maximize the expected number of clicks as a proxy of these platform's performance. As such, the study of click-through rate (CTR) prediction of advertisements, in environments like online social media, is of much interest. Prior works build machine learning (ML) using user-specific data to classify whether a user will click on an advertisement or not. For our work, we consider a large set of Facebook advertisement data (with no user data) and categorize targeted interests into thematic groups we call conceptual nodes. ML models are trained using the advertisement data to perform CTR prediction with conceptual node combinations. We then cast the problem of finding the optimal combination of conceptual nodes as an optimization problem. Given a certain budget k, we are interested in finding the optimal combination of conceptual nodes that maximize the CTR. We discuss the hardness and possible NP-hardness of the optimization problem. Then, we propose a greedy algorithm and a genetic algorithm to find near-optimal combinations of conceptual nodes in polynomial time, with the genetic algorithm nearly matching the optimal solution. We observe that simple ML models can exhibit the high Pearson correlation coefficients w.r.t. click predictions and real click values. Additionally, we find that the conceptual nodes of “politics”, “celebrity”, and “organization” are notably more influential than other considered conceptual nodes.
-
“Situational awareness using edge-computing enabled Internet-of-Things for smart grids”
Md Abul Hasnat, Md Jakir Hossain, Adetola Adeniran, Mahshid Rahnamay-Naeini, Hana Khamfroush • IEEE GLOBECOM Workshops • 2019
-
“Meeting Users’ QoS in a Edge-to-Cloud Platform via Optimally Placing Services and Scheduling Tasks”
Matthew Turner, Hana Khamfroush • IEEE ICNC CNC Workshop • 2020
-
“Reliable and Efficient Mobile Edge Computing for Dynamic IoT Systems”
Minoo Hosseinzadeh, Hana Khamfroush • ACM/IEEE SEC PhD Forum • 2019
-
“A Proximity-Based Generative Model for Online Social Network Topologies”
Emory Hufbaer, Nathaniel Hudson, Hana Khamfroush • IEEE ICNC • 2020
-
“Service Placement and Request Scheduling for Data-intensive Applications in Edge Clouds”
Vajiheh Farhadi, Fidan Mehmeti, Thomas La Porta, Ting He, Hana Khamfroush, Shiqiang Wang, Kevin S. Chan • IEEE INFOCOM • 2019
-
“On the effectiveness of standard centrality metrics for interdependent networks”
Nathaniel Hudson, Matthew Turner, Asare Nkansah, Hana Khamfroush • IEEE ICNC • 2019
-
“QoS-Aware Service Placement and Task Scheduling in a Three-Tiered Edge-to-Cloud System”
Matthew Turner, Hana Khamfroush • ACM/IEEE SEC PhD Forum • 2018
-
“Vulnerability of Interdependent Infrastructures Under Random Attacks”
Hana Khamfroush, Samuel Iloo, Mahshid Rahnamay-Naeini • ACM SIGMETRICS Workshops • 2018
-
“It’s Hard to Share: Joint Service Placement and Request Scheduling in Edge Clouds with Sharable and Non-sharable Resources”
Ting He, Hana Khamfroush, Shiqiang Wang, Tom La Porta, Sebastian Stein • IEEE ICDCS • 2018
-
“Controlling Cascading Failures in Interdependent Networks Under Uncertain Knowledge of Damage”
Diman Zad Tootaghaj, Novella Bartolini, Hana Khamfroush, Thomas La Porta • IEEE SRDS • 2017
-
“Network Recovery from Massive Failures under Uncertain Knowledge of Damages”
Diman Zad Tootaghaj, Hana Khamfroush, Novella Bartolini, Stefano Ciavarella, Seamus Hayes, Thomas La Porta • IFIP Networking • 2017
-
“Fundamental Limits of Failure Identiability by Boolean Network Tomography”
Novella Bartolini, Ting He, Hana Khamfroush • IEEE INFOCOM • 2017
-
“Progressive Damage Assessment and Network Recovery after Massive Failures”
Stefano Ciavarella, Novella Bartolini, Hana Khamfroush, Thomas La Porta • IEEE INFOCOM • 2017
-
“Service Placement for Detecting and Localizing Failures Using End-to-End Observations”
Ting He, Novella Bartolini, Hana Khamfroush, Injung Kim, Liang Ma, Thomas La Porta • IEEE ICDCS • 2016
-
“Relay-assisted Network Coding Multicast in the Presence of Neighbors”
Hana Khamfroush, Daniel E. Lucani, Peyman Pahlevani, Frank H. P. Fitzek, João Barros • VDE European Wireless • 2015
-
“Network coding for wireless cooperative networks: Simple rules, near-optimal delay”
Hana Khamfroush, Daniel E. Lucani, João Barros • IEEE ICC • 2014
-
“On the coded packet relay network in the presence of neighbors: Benefits of speaking in a crowded room”
Hana Khamfroush, Peyman Pahlevani, Daniel E. Lucani, Martin Hundebøll, Frank H. P. Fitzek • IEEE ICC • 2014
-
“Minimizing The Completion Time Of A Wireless Cooperative Network Using Network Coding”
Hana Khamfroush, Daniel E. Lucani, João Barros • IEEE PIMRC • 2013
Abstract
We consider the performance of network coding for a wireless cooperative network in which a source wants to transmit M data packets to two receivers. We assume that receivers can share their received packets with each other or simply wait to receive the packets from the source. The problem of finding an optimum packet transmission policy that minimizes the completion time in such a network is solved by modeling the problem as a Markov Decision Process (MDP). Our analysis is useful for a series of network coding and forwarding schemes with or without feedback. Our results show that the optimal network coding solution in terms of completion time, outperforms broadcasting with network coding by a factor of 2.13 and outperforms forwarding mechanisms by a factor of 6.1. Beyond computing the optimal completion time, we identify the critical decision policies derived from the MDP solution.
-
“GeoCode: A geographic coding-aware communication protocol”
Hana Khamfroush, Daniel E. Lucani, João Barros • IEEE ITSC • 2011
-
26 papers