Securing Machine Learning in the Cloud: A Systematic Review of Cloud Machine Learning Security

Oct 25, 2020·
Adnan Qayyum
,
Aneeqa Ijaz
,
Muhammad Usama
Dr. Waleed Iqbal
Dr. Waleed Iqbal
,
Junaid Qadir
,
Yehia Elkhatib
,
Ala Al-Fuqaha
· 0 min read
Abstract
Machine Learning as a Service (MLaaS) cloud platforms are widely used, and outsourcing of DL model training to third-party cloud services is increasing. This widespread usage opens many attack surfaces. We conduct a systematic review of literature on cloud-hosted ML/DL models along attacks and defenses dimensions: 31 articles, 19 attack-focused, 6 defense-focused, 6 both. We highlight limitations, pitfalls, and open research issues for cloud-hosted ML security.
Type
Publication
Frontiers in Big Data, 3:587139
publications
Machine Learning Security Review
Dr. Waleed Iqbal
Authors
Dr. Waleed Iqbal

I am Waleed Iqbal, an Assistant Professor in Data Science at Northeastern University, based at their London Campus.

I also hold positions of Teaching Fellow in Computer Science at the School of Electronic Engineering and Computer Science, Queen Mary University of London and Associate Lecturer at Arden University London.

I received my PhD in Computer Science under the supervision of Dr. Ignacio Castro and Prof. Gareth Tyson in the Social Data Science (SDS) Lab, Networks Research Group at Queen Mary University of London.

My research interests are broadly in manifestation of socio-economic inequality in online user activity.

Previously, I worked at IHSAN Lab in Information Technology University Lahore, Pakistan, supervised by Prof. Dr. Junaid Qadir as a postgraduate thesis student, affiliated researcher, and teaching assistant.

Since September 2023, I am endorsed and recognised as UK Global Talent (Exceptional Promise) in the Research and Academic Category by the Royal Academy of Engineering UK and granted Global Talent Visa by the Government of the United Kingdom.