Shahbaz Rezaei
YOU?
Author Swipe
View article: Domain Generalization in-the-Wild: Disentangling Classification from Domain-Aware Representations
Domain Generalization in-the-Wild: Disentangling Classification from Domain-Aware Representations Open
Evaluating domain generalization (DG) for foundational models like CLIP is challenging, as web-scale pretraining data potentially covers many existing benchmarks. Consequently, current DG evaluation may neither be sufficiently challenging …
View article: FixCLR: Negative-Class Contrastive Learning for Semi-Supervised Domain Generalization
FixCLR: Negative-Class Contrastive Learning for Semi-Supervised Domain Generalization Open
Semi-supervised domain generalization (SSDG) aims to solve the problem of generalizing to out-of-distribution data when only a few labels are available. Due to label scarcity, applying domain generalization methods often underperform. Cons…
View article: On the Necessity of Multi-Domain Explanation: An Uncertainty Principle Approach for Deep Time Series Models
On the Necessity of Multi-Domain Explanation: An Uncertainty Principle Approach for Deep Time Series Models Open
A prevailing approach to explain time series models is to generate attribution in time domain. A recent development in time series XAI is the concept of explanation spaces, where any model trained in the time domain can be interpreted with…
View article: Genotype Prediction from Retinal Fundus Images Using Deep Learning in Eyes with Age-Related Macular Degeneration
Genotype Prediction from Retinal Fundus Images Using Deep Learning in Eyes with Age-Related Macular Degeneration Open
Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
View article: Explanation Space: A New Perspective into Time Series Interpretability
Explanation Space: A New Perspective into Time Series Interpretability Open
Human understandable explanation of deep learning models is essential for various critical and sensitive applications. Unlike image or tabular data where the importance of each input feature (for the classifier's decision) can be directly …
View article: Benchmarking Counterfactual Interpretability in Deep Learning Models for Time Series Classification
Benchmarking Counterfactual Interpretability in Deep Learning Models for Time Series Classification Open
The popularity of deep learning methods in the time series domain boosts interest in interpretability studies, including counterfactual (CF) methods. CF methods identify minimal changes in instances to alter the model predictions. Despite …
View article: A quantitative analysis of fidgeting in ADHD and its relation to performance and sustained attention on a cognitive task
A quantitative analysis of fidgeting in ADHD and its relation to performance and sustained attention on a cognitive task Open
Introduction Attention-Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder where hyperactivity often manifests as fidgeting, a non-goal-directed motoric action. Many studies demonstrate fidgeting varies under different c…
View article: Dynamic Batch Norm Statistics Update for Natural Robustness
Dynamic Batch Norm Statistics Update for Natural Robustness Open
DNNs trained on natural clean samples have been shown to perform poorly on corrupted samples, such as noisy or blurry images. Various data augmentation methods have been recently proposed to improve DNN's robustness against common corrupti…
View article: Uncovering the Gut–Liver Axis Biomarkers for Predicting Metabolic Burden in Mice
Uncovering the Gut–Liver Axis Biomarkers for Predicting Metabolic Burden in Mice Open
Western diet (WD) intake, aging, and inactivation of farnesoid X receptor (FXR) are risk factors for metabolic and chronic inflammation-related health issues ranging from metabolic dysfunction-associated steatotic liver disease (MASLD) to …
View article: Predicting antimicrobial resistance of bacterial pathogens using time series analysis
Predicting antimicrobial resistance of bacterial pathogens using time series analysis Open
Antimicrobial resistance (AMR) is arguably one of the major health and economic challenges in our society. A key aspect of tackling AMR is rapid and accurate detection of the emergence and spread of AMR in food animal production, which req…
View article: Machine Learning to Identify Molecular Markers for Metabolic Disease Development Using Mouse Models
Machine Learning to Identify Molecular Markers for Metabolic Disease Development Using Mouse Models Open
Background Aging, Western diet (WD) intake, and bile acid (BA) receptor farnesoid X receptor (FXR) inactivation are risk factors for metabolic disease development including nonalcoholic fatty liver disease (NAFLD) and chronic inflammation-…
View article: On the Discredibility of Membership Inference Attacks
On the Discredibility of Membership Inference Attacks Open
With the wide-spread application of machine learning models, it has become critical to study the potential data leakage of models trained on sensitive data. Recently, various membership inference (MI) attacks are proposed to determine if a…
View article: User-Level Membership Inference Attack against Metric Embedding Learning
User-Level Membership Inference Attack against Metric Embedding Learning Open
Membership inference (MI) determines if a sample was part of a victim model training set. Recent development of MI attacks focus on record-level membership inference which limits their application in many real-world scenarios. For example,…
View article: An Efficient Subpopulation-based Membership Inference Attack
An Efficient Subpopulation-based Membership Inference Attack Open
Membership inference attacks allow a malicious entity to predict whether a sample is used during training of a victim model or not. State-of-the-art membership inference attacks have shown to achieve good accuracy which poses a great priva…
View article: Applications of Machine Learning for the Classification of Porcine Reproductive and Respiratory Syndrome Virus Sublineages Using Amino Acid Scores of ORF5 Gene
Applications of Machine Learning for the Classification of Porcine Reproductive and Respiratory Syndrome Virus Sublineages Using Amino Acid Scores of ORF5 Gene Open
Porcine reproductive and respiratory syndrome is an infectious disease of pigs caused by PRRS virus (PRRSV). A modified live-attenuated vaccine has been widely used to control the spread of PRRSV and the classification of field strains is …
View article: Accuracy-Privacy Trade-off in Deep Ensemble: A Membership Inference Perspective
Accuracy-Privacy Trade-off in Deep Ensemble: A Membership Inference Perspective Open
Deep ensemble learning has been shown to improve accuracy by training multiple neural networks and averaging their outputs. Ensemble learning has also been suggested to defend against membership inference attacks that undermine privacy. In…
View article: Accuracy-Privacy Trade-off in Deep Ensembles.
Accuracy-Privacy Trade-off in Deep Ensembles. Open
Deep ensemble learning has been shown to improve accuracy by training multiple neural networks and fusing their outputs. Ensemble learning has also been used to defend against membership inference attacks that undermine privacy. In this pa…
View article: Multitask Learning for Network Traffic Classification
Multitask Learning for Network Traffic Classification Open
Traffic classification has various applications in today's Internet, from\nresource allocation, billing and QoS purposes in ISPs to firewall and malware\ndetection in clients. Classical machine learning algorithms and deep learning\nmodels…
View article: On the Difficulty of Membership Inference Attacks
On the Difficulty of Membership Inference Attacks Open
Recent studies propose membership inference (MI) attacks on deep models, where the goal is to infer if a sample has been used in the training process. Despite their apparent success, these studies only report accuracy, precision, and recal…
View article: Towards the Infeasibility of Membership Inference on Deep Models
Towards the Infeasibility of Membership Inference on Deep Models Open
Recent studies propose membership inference (MI) attacks on deep models. Despite the moderate accuracy of such MI attacks, we show that the way the attack accuracy is reported is often misleading and a simple blind attack which is highly u…
View article: Large-Scale Mobile App Identification Using Deep Learning
Large-Scale Mobile App Identification Using Deep Learning Open
Many network services and tools (e.g. network monitors, malware-detection\nsystems, routing and billing policy enforcement modules in ISPs) depend on\nidentifying the type of traffic that passes through the network. With the\nwidespread us…
View article: Security of Deep Learning Methodologies: Challenges and Opportunities
Security of Deep Learning Methodologies: Challenges and Opportunities Open
Despite the plethora of studies about security vulnerabilities and defenses of deep learning models, security aspects of deep learning methodologies, such as transfer learning, have been rarely studied. In this article, we highlight the se…
View article: An Ambulatory Blood Pressure Monitor Mobile Health System for Early Warning for Stroke Risk: Longitudinal Observational Study
An Ambulatory Blood Pressure Monitor Mobile Health System for Early Warning for Stroke Risk: Longitudinal Observational Study Open
Background Stroke, as a leading cause of death around the globe, has become a heavy burden on our society. Studies show that stroke can be predicted and prevented if a person’s blood pressure (BP) status is appropriately monitored via an a…
View article: An Ambulatory Blood Pressure Monitor Mobile Health System for Early Warning for Stroke Risk: Longitudinal Observational Study (Preprint)
An Ambulatory Blood Pressure Monitor Mobile Health System for Early Warning for Stroke Risk: Longitudinal Observational Study (Preprint) Open
BACKGROUND Stroke, as a leading cause of death around the globe, has become a heavy burden on our society. Studies show that stroke can be predicted and prevented if a person’s blood pressure (BP) status is appropriately monitored via an …
View article: Deep Learning for Encrypted Traffic Classification: An Overview
Deep Learning for Encrypted Traffic Classification: An Overview Open
Traffic classification has been studied for two decades and applied to a wide\nrange of applications from QoS provisioning and billing in ISPs to\nsecurity-related applications in firewalls and intrusion detection systems.\nPort-based, dat…
View article: A Target-Agnostic Attack on Deep Models: Exploiting Security Vulnerabilities of Transfer Learning
A Target-Agnostic Attack on Deep Models: Exploiting Security Vulnerabilities of Transfer Learning Open
Due to insufficient training data and the high computational cost to train a deep neural network from scratch, transfer learning has been extensively used in many deep-neural-network-based applications. A commonly used transfer learning ap…
View article: A Target-Agnostic Attack on Deep Models: Exploiting Security\n Vulnerabilities of Transfer Learning
A Target-Agnostic Attack on Deep Models: Exploiting Security\n Vulnerabilities of Transfer Learning Open
Due to insufficient training data and the high computational cost to train a\ndeep neural network from scratch, transfer learning has been extensively used\nin many deep-neural-network-based applications. A commonly used transfer\nlearning…
View article: How to Achieve High Classification Accuracy with Just a Few Labels: A Semi-supervised Approach Using Sampled Packets
How to Achieve High Classification Accuracy with Just a Few Labels: A Semi-supervised Approach Using Sampled Packets Open
Network traffic classification, which has numerous applications from security to billing and network provisioning, has become a cornerstone of today's computer networks. Previous studies have developed traffic classification techniques usi…
View article: Throughput Analysis of IEEE 802.11 Multi-Hop Wireless Networks With Routing Consideration: A General Framework
Throughput Analysis of IEEE 802.11 Multi-Hop Wireless Networks With Routing Consideration: A General Framework Open
The end-to-end throughput of multi-hop communication in wireless ad hoc networks is affected by the conflict between forwarding nodes. It has been shown that sending more packets than maximum achievable end-to-end throughput not only fails…
View article: Study of Uniform Jitter Mechanism for Metric-based Wireless Routing
Study of Uniform Jitter Mechanism for Metric-based Wireless Routing Open
Many wireless protocols wait a small and random amount of time which is called jitter before sending a packet to avoid high contention and packet collision. Jitter has been already proposed for many routing protocols including AODV and LOA…