About

I am a doctoral student in the Data-driven Network and cyber Security (DANS) Lab at the University of Alberta, supervised by Dr. Euijin Choo and  Dr. Jörg Sander. My research focuses on using machine learning to improve computer security. Currently, I am working on building continual anomaly detection algorithms for real-time security data. Also, I am currently a research intern at Mastercard, working on improving fraud detection using graph neural networks by analyzing node interaction patterns.

I did my M.Sc at the University of Alberta under the supervision of Dr. Jörg Sander and Dr. Murilo Naldi. My M.Sc thesis studied the behaviour of GLOSH and proposed a direction to achieve unsupervised parameter less outlier detection. Previously, I studied the impact of data-driven and structural complexities in Deep Learning under the supervision of Dr. Nathalie Japkowicz. In my undergrad, I worked on data imbalance and overlapping problems in domains such as Sentiment Analysis, and Sarcasm Detection under the supervision of Dr. Sankhadeep Chatterjee. 

Want to talk about intrusion/anomaly detection? let's grab a coffee!

Education

• Ph.D in Computing Science, University of Alberta,  Jan 2024 - present

Thesis: Adaptive Provenance Intrusion Detection

Supervisors: Dr. Euijin Choo and Dr. Jörg Sander

• M.Sc in Computing Science, University of Alberta,  Sept 2021 - Jan 2024

Thesis: Beyond Defaults: Parameter Free Outlier Detection using GLOSH  (link, pdf, slides)

Supervisors: Dr. Jörg Sander and Dr. Murilo Naldi

Email: kushanku@ualberta.ca

Links: LinkedIn, Google Scholar, GitHub, ResearchGate, X

Goals. I am interested in how intelligent systems can learn, adapt, and reason about evolving environments. I explore these questions through cybersecurity, building systems that can learn complex system behavior, adapt to change, and anticipate evolving cyber threats.

More about me? (Personal)

"I build systems to solve complex problems in the real world"

Born?: I was born in the summer of '99 (and yes, Summer Of '69 by Bryan Adams is one of my favorite songs)

What fascinates me? (apart from Machine Learning): I'm fascinated by the concept of time and paradoxes. I love exploring topics like black holes and the origins of the universe

Books?: I enjoy both fiction and non-fiction, especially crime thrillers and unsolved mysteries. Agatha Christie is one of my favorite authors

Music?: I'm a music lover. I generally listen to pop, soft rock, country, and film scores

Religious views?: Atheist

One Dream?: To walk on the Moon! It may be funny and unrealistic, but hey, dreams are meant to be big ;-)

Interesting papers to read

[-] Duan et al. Breaking the Sorting Barrier for Directed Single-Source Shortest Paths, STOC 2025

[-] Li et al. More Agents Is All You Need, TMLR 2024

[-] Schaeffer et al. Are Emergent Abilities of Large Language Models a Mirage?, NeurIPS 2023

[-] Wei et al. Emergent abilities of large language models, TMLR 2022

[-] Bagdasaryan et al. Differential privacy has disparate impact on model accuracy, NeurIPS 2019

[-] Schubert et al. DBSCAN revisited, revisited: why and how you should (still) use DBSCAN, ACM TODS 2017