Project Start Date:
September 2016
Project Status:
On-going
Project Objectives and Scope
The main project goal is to develop an infrastructure that will collect information using agents in a fully trusted manner. The software agents will collect information from servers and end-points. The nature of the collected information will be determined with the help of design partners of STEE-Infosec.
Challenges
The main research challenge is how to isolate the monitoring infrastructure from the potentially compromised system that is being monitored. The software agent will have to run in a system that may be compromised.
Methodology/Approach
For implementing the agent we intend to use various virtualisation and trusted computing techniques that will help us to protect the agent from being compromised. We intend to activate the agents such that each agent will crosscheck the wellbeing of other agents running in other computers.
Publications
Adaptive Noise Injection for Training Stochastic Student Networks from Deterministic Teachers
Analysing the Adversarial Landscape of Binary Stochastic Networks
Simple and Effective Prevention of Mode Collapse in Deep One-Class Classification
Deep Semi-Supervised Anomaly Detection
Exploring the Back Alleys: Analysing The Robustness of Alternative Neural Network Architectures against Adversarial Attacks
A Neural Attention Model for Real-Time Network Intrusion Detection
Memory snapshot dataset of a compromised host with malware using obfuscation evasion techniques
The DUSTER Attack: Tor Onion Service Attribution Based on Flow Watermarking with Track Hiding
Adversarial Attacks on Remote User Authentication Using Behavioural Mouse Dynamics
User Authentication Based on Mouse Dynamics Using Deep Neural Networks: A Comprehensive Study
Nonintrusive heart rate measurement using ballistocardiogram signals: a comparative study
Unmasking Clever Hans predictors and assessing what machines really learn
Deep One-Class Classification
Mouse Authentication without the Temporal Aspect – What does a 2D-CNN learn?
Insights from curve fitting models in mouse dynamics authentication systems
ImageCLEF 2017: ImageCLEF Tuberculosis Task-the SGEast Submission
Principal Investigator(s)

Professor ELOVICI, Yuval
- yuval_elovici@sutd.edu.sg
- +65 6499 4919
-
- Trusted and Resilient Monitoring Infrastructure (Principal Investigator)

CHENG, Tai Leong Jimmy
Co-Principal Investigator(s)

CHEUNG, Ngai-Man (Man)
- ngaiman_cheung@sutd.edu.sg
- +65 6499 4542
- 1.502.17
- SUTD Profile
-
- Big Data Security Analytics (Principal Investigator)
- Predicting Adversarial Behaviours and the Motivation for Automated Network Defense (Co-Principal Investigator)
- Trusted and Resilient Monitoring Infrastructure (Co-Principal Investigator)

PILIOURAS, Georgios
- georgios@sutd.edu.sg
- +65 6499 4545
- 1.702.09
- SUTD Profile
-
- Trusted and Resilient Monitoring Infrastructure (Co-Principal Investigator)
Researcher(s)

IACOVAZZI, Alfonso
- alfonso_iacovazzi@sutd.edu.sg
- 2.301
-
- Securing Avionics Communication Bus (Principal Investigator)
- Trusted and Resilient Monitoring Infrastructure (Researcher)