BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Electrical & Computer Engineering - ECPv6.16.2//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Department of Electrical & Computer Engineering
X-ORIGINAL-URL:https://ece.northeastern.edu
X-WR-CALDESC:Events for Department of Electrical & Computer Engineering
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20210314T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20211107T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20220313T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20221106T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20230312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20231105T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20221202T080000
DTEND;TZID=America/New_York:20221202T170000
DTSTAMP:20260617T182928
CREATED:20220824T182336Z
LAST-MODIFIED:20220824T182336Z
UID:5782-1669968000-1670000400@ece.northeastern.edu
SUMMARY:First Year Engineering Expo
DESCRIPTION:Please come to the Curry Student Center indoor quad and pit on Friday\, December 2nd to see Northeastern’s First-Year Engineering Students’ inventive projects\, games\, and exhibits. \nStudents will showcase original board games\, interactive projects geared to teach children sustainability concepts\, and prolific prototypes to help solve a wide range of problems. \nEach project applies the engineering concepts introduced this past semester\, which includes the Engineering Design Process\, Solidworks\, AutoCAD\, Programming with C++ and Matlab\, and controlling microelectronics with Arduino.
URL:https://ece.northeastern.edu/event/first-year-engineering-expo-3/
LOCATION:Curry Student Center\, 360 Huntington Ave.\, Boston\, MA\, 02115\, United States
GEO:42.3394629;-71.0885286
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Curry Student Center 360 Huntington Ave. Boston MA 02115 United States;X-APPLE-RADIUS=500;X-TITLE=360 Huntington Ave.:geo:-71.0885286,42.3394629
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20221205T100000
DTEND;TZID=America/New_York:20221205T110000
DTSTAMP:20260617T182928
CREATED:20221201T022546Z
LAST-MODIFIED:20221201T022546Z
UID:6004-1670234400-1670238000@ece.northeastern.edu
SUMMARY:Ramtin Khalili's PhD Dissertation Defense
DESCRIPTION:Abstract: \nState estimation is a critical application in energy management systems. Due to the increased penetration of inverter-based resources\, installed advanced infrastructure at all voltage levels\, and unconventional loads like electric vehicle charging stations\, a three-phase state estimator formulation is essential. The first issue is the convoluted formulation and modeling techniques that are required in three-phase systems studies. Moreover\, the size of network matrices expanded\, which makes the analysis computationally costly. This dissertation addresses this by proposing a new decoupled state estimation method. The idea is to exploit the linearity of measurement equations\, decompose the three-phase coupled equations into three independent modal measurement equations\, perform the state estimation independently for each mode\, and finally reconstruct the three-phase quantities. This method is applicable to both radial and meshed three-phase networks. Furthermore\, multi-phase structures can be handled by the new estimator\, which makes the approach practical when monitoring mixed-phase feeder sections is of interest. \nWhile utilities are investing in expanding the grid and installing more PMUs\, there might not be enough PMUs to make the network observable in all networks\, especially at lower voltage levels. So\, PMU-based linear state estimators are not always feasible. On the other hand\, SCADA measurements are available with adequate redundancy in most networks. However\, SCADA-based state estimation is nonlinear\, which brings various problems like divergence issues and significant CPU times. The computational complexity will be even worse if the three-phase state estimation is formulated based on SCADA measurements due to their nonlinear nature\, which makes modal decoupling impossible. So\, a new linear formulation has been proposed for both the positive-sequence and three-phase networks based on conventional measurements. This approach converts the nonlinear recursive problem into an iterative linear state estimation problem. \nThe inherent assumption in most of the state estimators is a perfect network model. However\, network parameter errors are susceptible to errors that can bias the state estimation solution. This can deceive the existing bad data tools as parameter errors appear as if multiple interacting measurement errors occur locally. So\, a two-stage method is proposed for parameter error identification and correction for large three-phase networks. A systematic PMU placement strategy is also proposed to ensure the detectability of parameter errors. The benefits of multi-area state estimation are demonstrated for the deregulated power grids for monitoring the local and boundary areas. It has also shown promising results in increasing the efficiency of state estimation using a distributed framework. Parameter and measurement errors can remain undetected as a result of weakened measurement redundancy on the boundaries. However\, boundary errors in the area boundaries will be detected due to measurement consolidation at the coordination level. \nCommittee:\nProf. Ali Abur (Advisor)\nProf. Bahram Shafai\nProf. Mahshid Amirabadi
URL:https://ece.northeastern.edu/event/ramtin-khalilis-phd-dissertation-defense/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20221206T160000
DTEND;TZID=America/New_York:20221206T173000
DTSTAMP:20260617T182928
CREATED:20221206T020005Z
LAST-MODIFIED:20221206T020005Z
UID:6017-1670342400-1670347800@ece.northeastern.edu
SUMMARY:Md Navid Akbar's PhD Dissertation Defense
DESCRIPTION:“Inference from Brain Imaging: Incorporating Domain Knowledge and Latent Space Modeling” \nAbstract:\n\nBrain imaging can probe the anatomy (structural) of our brain\, or its function (functional). A particular imaging modality (unimodal) generally provides only a particular insight into human health. Transcranial magnetic stimulation (TMS)\, though still in its infancy as a brain imaging modality\, is such a functional\, unimodal technique. TMS helps model human motor-cortical mapping\, using corresponding muscle activity captured by surface electromyography (EMG)\, but it necessitates a reliable data-driven model. Earlier works have modeled the causal direction only (from cortical representation to muscles)\, or the inverse direction (from muscles to cortical representation)\, with simple statistical regression. We modeled this motor-cortical mapping bi-directionally in this dissertation\, using deep learning. We first modeled TMS-induced 3D electric field (E-field) in a brain to causal multi-muscle activation picked up by EMG\, in a regression task using a convolutional neural network (CNN) autoencoder. By fusing neuroscience domain knowledge (e.g.\, an empirical neural response profile)\, we reduced 14% squared error\, compared to the baseline model that did not contain this. We then designed our novel inverse imaging CNN model\, to reconstruct physiologically meaningful E-field distributions (in the image domain) from a given set of muscle activations (in the sensor domain). By adopting variational inference in the CNN model\, to learn the underlying latent space better\, we were able to reduce 13% in squared error over our purely CNN baseline. \nDiagnosis with brain imaging is often incomplete with a unimodal technique\, and having multiple sources (multimodal) may be advantageous. Successful multimodal fusion can provide more holistic information\, compared to its constituents. One relevant example is the classification of late post-traumatic seizure (LPTS). Previous works in this space have tackled LPTS classification with either unimodal functional imaging\, or non-machine learning (ML) structural modeling. In this dissertation\, we first undertook the ML classification of binary LPTS: with unimodal\, structural brain imaging\, namely diffusion magnetic resonance imaging (dMRI). By incorporating interpretable domain knowledge (post-traumatic lesion volume compensation)\, we improved 7% in the mean area under the curve (AUC) over the standard technique in literature. Finally\, we classified LPTS for a larger sample of subjects\, utilizing multimodal imaging\, including functional MRI (fMRI) and electroencephalography (EEG). Following unsupervised imputation for any missing modality within the subjects\, we introduced our novel multimodal fusion algorithm\, which attempts to leverage the underlying structure of the multivariate information. We found that our proposed algorithm improved by 7% in AUC performance\, over a naive Bayesian estimator that can handle missing data intrinsically.\nCollectively\, the work presented here demonstrated that incorporating domain knowledge in the modeling pipeline successfully improved inference. Similar improvements were also observed by learning and leveraging the possible underlying latent structure of the given information\, and adapting the models accordingly. \n\n\n\nCommittee:\n\nProf. Deniz Erdogmus (Advisor) \nProf. Mathew Yarossi (Co-advisor)\nProf. Dominique Duncan\nProf. Sarah Ostadabbas
URL:https://ece.northeastern.edu/event/md-navid-akbars-phd-dissertation-defense/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20221208T110000
DTEND;TZID=America/New_York:20221208T120000
DTSTAMP:20260617T182928
CREATED:20221201T023045Z
LAST-MODIFIED:20221201T023045Z
UID:6008-1670497200-1670500800@ece.northeastern.edu
SUMMARY:Danlin Jia's PhD Dissertation Defense
DESCRIPTION:“Towards Performance and Cost-efficiency for Data-intensive Applications in Distributed Data Processing Systems” \nAbstract: \nData-intensive science (DIS) has experienced a significant boom in the past decade. The emerging technologies of data-intensive services and infrastructures contribute to DIS’s development and raise challenges. An ecosystem has been constructed considering performance\, scalability\, sustainability\, and reliability to provide a high-quality service to DIS applications. The ecosystem consists of services exposed to users for application deployment and infrastructures to support data storage\, transfer\, and management from the system’s perspective. DIS applications share typical features\, such as memory and I/O intensity. Thus\, addressing the bottlenecks triggered by memory-intensive or I/O-intensive workloads in services and infrastructures is essential to improve the performance and cost-efficiency of the whole ecosystem. In this dissertation\, we investigate the characteristics of various DIS applications and design new resource allocation and scheduling schemes for the services and infrastructures in the DIS ecosystem. \nWe first investigate memory optimization in DIS ecosystems. In-memory data analytic frameworks are proposed to cache critical intermediate data in memory instead of in storage drives. Apache Spark is a commonly adopted in-memory data analytic framework with two memory managers\, Static and Unified. However\, the static memory manager lacks flexibility. In contrast\, the unified memory manager puts heavy pressure on the garbage collection of the Java Virtual Machine on which Spark resides. To address these issues\, we propose a new learning-based bidirectional usage-bounded memory allocation scheme to support dynamic memory allocation considering both memory demands and latency introduced by garbage collection. Distributed data-processing workloads in container-based virtualization take advantage of resource sharing\, fast delivery\, and excellent portability of containerization but also suffer from resource competition and performance interference. This inevitably induces performance degradation and significantly long latency\, even worse when over-provisioning. Motivated by this problem\, we design an efficient memory allocation scheme (RITA) for containerized parallel systems to improve data processing latency. RITA monitors applications’ memory usage and cache characteristics and dynamically re-allocates memory resources. \nWe also propose I/O optimizations for DIS applications and infrastructures. Distributed Deep Learning (DDL) accelerates DNN training by distributing training workloads across multiple computation accelerators\, e.g.\, GPUs. Although a surge of research has been devoted to optimizing DDL training\, the impact of data loading on GPU usage and training performance has been relatively under-explored. When multiple DDL applications are deployed\, the lack of a practical and efficient technique for data-loader allocation incurs GPU idleness and degrades the training throughput. In this dissertation\, we thus investigate the impact of data-loading on the global training throughput and design a resource allocator that uses the data-loading rate as a knob to reduce the GPU idleness. Finally\, designs and optimizations on disaggregated storage systems supported by cutting-edge storage and network techniques emerge dramatically. Disaggregated storage systems can scale resources independently and provide high-quality services for hyper-scale architectures. The traditional congestion control mechanism relieves congestion by limiting the data-sending rate of senders. However\, such a design scarifies the storage drive’s performance as data are generated but stalled on storage host nodes if network congestion happens. To solve this issue\, we design a storage-side rate control mechanism to mitigate network congestion while avoiding sacrificing I/O performance. \nCommittee: \nProf. Ningfang Mi (Advisor) \nProf. Xue Lin \nProf. David Kaeli
URL:https://ece.northeastern.edu/event/danlin-jias-phd-dissertation-defense/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20221208T140000
DTEND;TZID=America/New_York:20221208T160000
DTSTAMP:20260617T182928
CREATED:20221202T201226Z
LAST-MODIFIED:20221202T201226Z
UID:6014-1670508000-1670515200@ece.northeastern.edu
SUMMARY:Chuangtang Wang's PhD Proposal Review
DESCRIPTION:“All-optical Control of Magnetization in Nanostructures” \nCommittee: \nProf. Yongmin Liu (Advisor) \nProf. Don Heiman \nProf. Nian X. Sun \nAbstract:\nThe switching of magnetization by a femtosecond laser within several picoseconds has recently gained substantial attention\, because it promises next-generation\, energy-efficient\, and high-rate data storage technology. One of the most intriguing demonstrations is the helicity-dependent switching (HD-AOS) of a ferromagnet\, in which the magnetization states can be deterministically written and erased using left- and right-circularly polarized light. However\, the challenge is to realize a single-pulse HD-AOS. Controlling the spin angular momentum transfer from light to magnetic materials in nanostructures is the key to advance this field.\nIn my thesis research work\, I will study the all-optical control of magnetization in different nanostructures\, aiming to better understand the underlying mechanisms of HD-AOD and accelerate the technology development. Firstly\, helicity-driven magnetization dynamics in heavy metal/ferromagnet Au(Pt)/Co bilayer by the optical spin transfer torque (OSTT) is experimentally explored. The wavelength-dependent measurement of OSTT reveals that the quantum efficiency of OSTT strongly depends on the interface electronic structure and pump energy. The Inverse Faraday effect (IFE)\, which is believed to be the driving mechanism of HD-AOS\, is subsequently investigated in an Au thin film. The dependence of IFE on photon energy implies that the orbital angular momentum contribution to IFE is dominated by the excitation of laser pulses. To the best of our knowledge\, it is the first demonstration of this phenomenon. Lastly\, I will discuss our recent results on plasmonics-enhanced all-optical control of magnetization. Light can be tightly confined in plasmonic structures\, which can potentially enable low-energy and high-density magnetic data storage.
URL:https://ece.northeastern.edu/event/chuangtang-wangs-phd-proposal-review/
LOCATION:138 ISEC\, 360 Huntington Ave\, 138 ISEC\, Boston\, MA\, 02115\, United States
GEO:42.3401758;-71.0892797
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=138 ISEC 360 Huntington Ave 138 ISEC Boston MA 02115 United States;X-APPLE-RADIUS=500;X-TITLE=360 Huntington Ave\, 138 ISEC:geo:-71.0892797,42.3401758
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20221209T110000
DTEND;TZID=America/New_York:20221209T130000
DTSTAMP:20260617T182928
CREATED:20221201T023204Z
LAST-MODIFIED:20221201T023204Z
UID:6010-1670583600-1670590800@ece.northeastern.edu
SUMMARY:Bin Sun's PhD Dissertation Defense
DESCRIPTION:“Factorization guided Lightweight Neural Networks for Visual Analysis” \nCommittee: \nProf. Yun Fu (Advisor) \nProf. Ming Shao \nProf. Lili Su \nAbstract: \nDeep learning has become popular in recent years primarily due to powerful computing devices such as GPUs. However\, many applications such as face alignment\, image classification\, and gesture recognition need to be deployed to multimedia devices\, smartphones\, or embedded systems with limited resources. Thus\, there is an urgent need for high-performance but memory-efficient deep learning models. For this\, we design several lightweight deep learning models for different tasks with factorization strategies. \nSpecifically\, we constructed a lightweight face alignment model by proposing a factorization-based deep convolution module named Depthwise Separable Block (DSB) and a light but practical module based on the spatial configuration of the faces. Experiments on four popular datasets verify that Block Mobilenet has better overall performance with less than 1MB storage size.\nBesides the face analysis application\, we also explored a general\, lightweight deep learning module for image classification with low-rank pointwise residual (LRPR) convolution\, called LRPRNet. Essentially\, LRPR aims at using a low-rank approximation to factorize the pointwise convolution while keeping depthwise convolutions as the residual module to rectify the LRPR module. Moreover\, our LRPR is quite general and can be directly applied to many existing network architectures. \nDue to the success of the factorization strategy on image-based data\, we extended factorization on time sequence data for Sign Language Recognition (SLR). We achieved the first rank in the challenge of SLR with the help of our proposed novel Separable Spatial-Temporal Convolution Network (SSTCN)\, which divides a 3D convolution on joint features into several stages \, which help the SSTCN achieve higher accuracy with fewer parameters. \nWe also tried to factorize the features for single image super resolution (SISR). Factorization on features will reduce the feature size in order to reduce the computation costs. However\, the reduction of the spatial size is counter-intuitive for the super resolution task. With our exploration\, we demonstrated a network named Hybrid Pixel-Unshuffled Network (HPUN)\, which factorized the features to achieve the lightweight purpose while keeping high performance. Specifically\, we utilized pixel-unshuffle operation to factorize the input features. After the factorization\, we improved the performance by the grouped convolution\, max-pooling\, and self-residual. The experiments on popular benchmarks showed that the factorization strategy could achieve SOTA performance on SISR.
URL:https://ece.northeastern.edu/event/bin-suns-phd-dissertation-defense/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20221209T120000
DTEND;TZID=America/New_York:20221209T133000
DTSTAMP:20260617T182928
CREATED:20221201T022737Z
LAST-MODIFIED:20221201T022840Z
UID:6006-1670587200-1670592600@ece.northeastern.edu
SUMMARY:Alexey Tazin's PhD Dissertation Defense
DESCRIPTION:“Composition of UML Class Diagrams Using Category Theory and External Constraints” \nAbstract:\nIn large software development projects there is always a need for refactoring and optimization of the design. Usually software designs are represented using UML diagrams (e.g class diagrams). A software engineering team may create multiple versions of class diagrams satisfying some external constraints. In some cases\, subdiagrams of the developed diagrams can be selected and combined into one diagram. It is difficult to perform this task manually since manual process is very time consuming\, is prone to human errors\, and is not manageable for large projects. In this dissertation we present an algorithmic support for automating the generation of composed diagrams\, where the composed diagram satisfies a given collection of external constraints and is optimal with respect to a given objective function. The composition of diagrams is based on the colimit operation from category theory. The developed approach was verified experimentally by generating random external constraints (expressed in SPARQL and OWL)\, generating random class diagrams using these external constraints\, generating composed diagrams that satisfy these external constraints\, and computing class diagram metrics for each composed diagram. \nCommittee: \nProf. Mieczyslaw Kokar (Advisor) \nProf. David Kaeli \nDr. Jeff Smith
URL:https://ece.northeastern.edu/event/alexey-tazins-phd-dissertation-defense/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20221215T090000
DTEND;TZID=America/New_York:20221215T110000
DTSTAMP:20260617T182928
CREATED:20221213T011124Z
LAST-MODIFIED:20221213T011124Z
UID:6023-1671094800-1671102000@ece.northeastern.edu
SUMMARY:Daniel Uvaydov's PhD Proposal Review
DESCRIPTION:“Real-Time Spectrum Sensing for Inference and Control”\n\nAbstract:\nSpectrum sensing can enable the next generation of wireless applications ranging from opportunistic spectrum access to cognitive radio networks. The key unaddressed challenges of spectrum sensing are that (i) it has to be performed with extremely low latency over varying bandwidths and must guarantee strict real-time processing constraints; (ii) its underlying algorithms need to be extremely accurate\, and flexible enough to work with different wireless bands and protocols to find application in real-world settings. We address these challenges in multiple wireless applications by utilizing Deep Learning techniques as the main vehicle of spectrum sensing for both inference and control. By leveraging mechanisms such as data augmentation\, channel attention\, voting\, and segmentation we are able to push beyond the capabilities of existing Deep Learning techniques and create generalizable spectrum sensing algorithms. Furthermore we deploy different spectrum sensing solutions in real testbeds for over the air evaluations and applicable proof-of-concepts.\n\n\nCommittee:\n\nProf. Tommaso Melodia (Advisor) \nProf. Francesco Restuccia\nProf. Kaushik Chowdhury
URL:https://ece.northeastern.edu/event/daniel-uvaydovs-phd-proposal-review/
LOCATION:432 ISEC\, 360 Huntington Ave\, Boston\, MA\, 02115\, United States
GEO:42.3396156;-71.0886534
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=432 ISEC 360 Huntington Ave Boston MA 02115 United States;X-APPLE-RADIUS=500;X-TITLE=360 Huntington Ave:geo:-71.0886534,42.3396156
END:VEVENT
END:VCALENDAR