Mina Jaberi

I am a Research Assistant (Ph.D. Candidate in Telecommunications) at INRS-EMT, University of Quebec supervised by Professor Tiago Falk working on physiological signal processing and Machine Learning. I am part of the Multisensory Signal Analysis and Enhancement (MuSAE) Lab. Currently I am conducting research on security and privacy concerns of the metaverse.

I have experience with several programming languages such as Python and .Net. I have 5+ years of experience as a Back-end and Database Developer working with relational databases. Additionally, I have 4+ years of research experience in Machine and Deep Learning. I have experience working both individually and in teams.

I received my M.Sc. in Computer Engineering in 2020. During my Master's degree, I worked on human activity recognition using Deep Learning. More specifically, I worked on an approach to be able to utilize Deep Learning for human activity recognition in wearable devices that have restricted computational and energy resources. I am interested in Applied Deep Learning and Machine Learning especially in healthcare.

Let's talk about fun stuff! I enjoy coffee and tea ☕. I like baking cookies and cake 🍪. I have been doing Yoga for years 🧘🏻‍♀️. And of course, I follow tech-related topics 💻📱!

Education

Institut National de la Recherche Scientifique (INRS)

2023-Present
Montreal, QC, Canada

Ph.D. in Telecommunications

Member of MuSAE AI Research Lab (MuSAE Lab)
Research Focus: Security and Privacy Concerns in the Metaverse

Islamic Azad University - Central Tehran Branch

2016-2020
Tehran, Iran

M.Sc. in Computer Engineering

Focused on Using Deep Learning for Human Activity Recognition
Thesis Title: Human Activity Recognition through wearable devices using deep neural networks (Published Paper)

Islamic Azad University - Central Tehran Branch

2010-2015
Tehran, Iran

BS in Computer Engineering

Selected Publications

2024

Biosensor-Instrumented xR Headsets: A Double-Edged Sword for User Identity and Privacy Management in the Metaverse

Ihsân Grichi, Mina Jaberi, Tiago H. Falk

Proceedings of the IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct)

View

2022

Human activity recognition via wearable devices using enhanced ternary weight convolutional neural network

Mina Jaberi, Reza Ravanmehr

Pervasive and Mobile Computing

View

@article{JABERI2022101620,
    title = {Human activity recognition via wearable devices using enhanced ternary weight convolutional neural network},
    journal = {Pervasive and Mobile Computing},
    volume = {83},
    pages = {101620},
    year = {2022},
    issn = {1574-1192},
    doi = {https://doi.org/10.1016/j.pmcj.2022.101620},
    url = {https://www.sciencedirect.com/science/article/pii/S1574119222000554},
    author = {Mina Jaberi and Reza Ravanmehr},
    keywords = {Human activity recognition, Deep neural network, Convolutional neural network, Ternary weight network, Wearable devices, Ubiquitous computing},
    abstract = {Human activity recognition (HAR) has been known as an active area for more than a decade, and there are still crucial aspects that are intended as challenging problems. Providing detailed and appropriate information about the activities and behaviors of people is one of the most important fields in ubiquitous computing. There are numerous applications in this field, among which healthcare, security, and entertainment scenarios can be listed. Human activity recognition can be carried out with the assistance of smartphone sensors such as accelerometers and gyroscopes or images captured from webcams. Today, the application of deep neural networks in this domain has received much attention and has led to more accurate and effective results compared to traditional techniques. The deep neural network performs arithmetic operations on a various number of hidden layers. In this article, a new approach called HAR-CT is proposed to enhance the accuracy of human activity recognition in various classes by adopting a convolutional neural network (CNN). Subsequently, an optimization technique using the TWN model is also suggested to reduce the complexity of the deep neural network approach that decreases the energy consumption of mobile devices. To this end, the float precision weights of the convolutional neural network are quantized and converted into ternary weights, while the decline in the accuracy is very low compared to the initial deep neural network. The evaluation results of both networks demonstrate that the proposed methods outperform the recently published approaches in human activity recognition.}
}

View All Publications