Dr. Nhien-An Le-Khac, Lecturer/Assistant Professor, School of Computer Science, University College Dublin, Ireland.

Profile:https://people.ucd.ie/an.lekhac

Biography: Dr. Nhien-An Le-Khac is a Lecturer (tenured) at the School of Computer Science (CS), UCD. He is currently the Programme Director of MSc programme in Forensic Computing and Cybercrime Investigation (FCCI, http://www.csi.ucd.ie/PostgraduateProgrammes/MSc_FCCI) – an international programme for the law enforcement officers specialising in cybercrime investigations. To date, more than 900 students from 62 countries in 5 continents have graduated from this FCCI programme. He is also the co-founder of UCD-GNECB Postgraduate Certificate in fraud and e-crime investigation (https://goo.gl/bSddQx). He was a Research Fellow in Citibank, Ireland (Citi). He obtained his PhD. in Computer Science in 2006 at the Institut National Polytechnique Grenoble (INPG), France. His research interest spans the area of Cybersecurity and Digital Forensics, Machine Learning/Distributed Data Mining for Security, Fraud and Criminal Detection, Cloud Security and Privacy, High Performance computing,and Secure Healthcare IT systems. Since 2013, he has collaborated on many research projects as a principal/co-PI/funded investigator. He has published more than 150 scientific papers in peer-reviewed journals and conferences in related research fields. He is also an active chair as well as a reviewer for many key conferences and journals in related disciplines.

Artificial Intelligence for Internet of Things Security and Forensics

Today is the era of Internet of Things (IoT), millions of machines such as cars, smoke detectors, watches, glasses, webcams, smart blood pressure monitor, etc. are being connected to the Internet. The number of machines that possess the ability of remote access to monitor and collect data is continuously increasing. This development makes, on one hand, the human life more comfortable, convenient, but it also raises on other hand issues on security, privacy. Some challenges can be listed as the multiple locations and networks; the management and automation features of an IoT network; the survival period and visibility nature of IoT devices; huge amount of IoT data collected. Due to the un-standardized nature of IoT devices, the cybercrimes can exploit security vulnerability of these devices as well as of their eco-systems such as the insecure web, backend API, cloud, or mobile interfaces, etc. Attacks such as sniffing, surveillance, DDoS, etc. on IoT networks and IoT devices are normally difficult to detect.

Recently, Artificial Intelligence (AI) have made tremendous progress in many areas such as computer vision, pattern recognition, natural language processing, cybersecurity, etc. Many AI-based approaches, using machine learning and deep learning models have been developed as a key solution to secure IoT eco-systems. Hence, in the first part of this talk, I am willing to present achievements of using machine learning and deep learning for IoT security and forensics.

On the other hand, such AI-based approaches themselves could also raise issues in privacy and security. It is more severe if these approaches are black-box to end-users or investigators. For example, deep learning models are trained on very large amounts of data from potentially untrustworthy sources, providing opportunities for adversaries to manipulate them. In recent years, many sophisticated adversarial attack techniques have been exploited to compromise deep learning based systems. Therefore, in the second part, I share my thought on how to provide the robustness and safety to AI-based models against these adversarial attacks.