Hassan Ali
PhD Candidate, UNSW

I am a Ph.D. candidate in Computer Science and Engineering (CSE) at the University of New South Wales (UNSW), Sydney. I am supervised by Dr. Salil Kanhere, Dr. Sanjay Jha and Dr. Surya Nepal.

My research interests are diverse.

• I want to enable ML as a product in real-world settings (e.g., ML as a service (supporting encryption standards/protocols), ML on edge).
• I am interested in few-shot media generation, particularly text-to-video generation, using AI.
• I want to build frameworks for reliable and trustworthy Artificial Intelligence (AI) algorithms focused on real-world applications and challenges.

My past research work has largely focused on the trustworthy Machine Learning (ML) algorithms, particularly Deep Neural Networks (DNNs), where the trustworthiness refers to the adversarial robustness, security, privacy, interpretability, alignment and fairness of DNNs.

I can be contacted at: hassan.ali@unsw.edu.au

News:

• Oct 2024: Our paper "Adversarially Guided Stateful Defense Against Backdoor Attacks in Federated Deep Learning" has been accepted for publication at Annual Computer Security Applications Conference (ACSAC) 2024.
  
• Aug 2024: Our paper "Robust Encrypted Inference in Deep Learning: A Pathway to Secure Misinformation Detection" has been accepted for publication at IEEE Transactions on Dependable and Secure Computing.
  
• Jul 2024: Our paper "R2S100K: Road-Region Segmentation Dataset For Semi-Supervised Autonomous Driving in the Wild" has been accepted for publication at International Journal of Computer Vision.
  
• Mar 2024: Our paper "Adversarial Machine Learning for Social Good: Reframing the Adversary as an Ally" has been accepted for publication at IEEE Transactions on Artificial Intelligence.
  
• Nov 2023: Our paper "Consistent Valid Physically-Realizable Adversarial Attack against Crowd-flow Prediction Models" has been accepted for publication at IEEE Transactions on Intelligent Transporation Systems.
  
• Sep 2023: I have joined UNSW as a PhD candidate. My research will focus on Trustwothy Machine Learning.
  
• Aug 2023: Our paper "Secure and Trustworthy Artificial Intelligence-Extended Reality (AI-XR) for Metaverses" has been accepted at ACM Computing Surveys.
  
• Jul 2023: Our paper "ConDetect: Detecting Adversarially Perturbed Natural Language Inputs to Deep Classifiers Through Holistic Analysis" has been accepted for publication at Computers & Security.
  
• Dec 2022: Our paper "Towards secure private and trustworthy human-centric embedded machine learning: An emotion-aware facial recognition case study" has been accepted for publication at Computers & Security.
  
• Jun 2022: Our paper "Tamp-X: Attacking Explainable Natural Language Classifiers Through Tampered Activations" has been accepted for publication at Computers & Secruity.
  

Selected Works:

1. Adversarially Guided Stateful Defense Against Backdoor Attacks in Federated Deep Learning
   Hassan Ali, Surya Nepal, Salil S. Kanhere, Sanjay Jha
   In 2024 Annual Computer Security Applications Conference (ACSAC).
   Links: [Paper] [Code]
   
   
2. Robust Encrypted Inference in Deep Learning: A Pathway to Secure Misinformation Detection
   Hassan Ali*, Rana Tallal Javed*, Adnan Qayyum, Amer AlGhadhban, Meshari Alazmi, Ahmad Alzamil, Khaled Al-utaibi, and Junaid Qadir (*equal contribution)
   IEEE Transactions on Dependable and Secure Computing (2024).
   Links: [Paper] [Code]
   
   
3. R2S100K: Road-Region Segmentation Dataset For Semi-Supervised Autonomous Driving in the Wild
   Muhammad Atif Butt, Hassan Ali, Adnan Qayyum, Waqas Sultani, Ala Al-Fuqaha, and Junaid Qadir
   International Journal of Computer Vision (2024).
   Links: [Paper] [Code]
   
   
4. Consistent Valid Physically-Realizable Adversarial Attack against Crowd-flow Prediction Models
   Hassan Ali, Muhammad Atif Butt, Fethi Filali, Ala Al-Fuqaha, and Junaid Qadir
   IEEE Transactions on Intelligent Transportation Systems (2023).
   Links: [Paper] [Code]
   
   
5. Con-detect: Detecting adversarially perturbed natural language inputs to deep classifiers through holistic analysis
   Hassan Ali*, Muhammad Suleman Khan*, Amer AlGhadhban, Meshari Alazmi, Ahmad Alzamil, Khaled AlUtaibi, and Junaid Qadir (*equal contribution)
   Computers & Security 132 (2023): 103367.
   Links: [Paper] [Code]