Currently, I am part of the SVR Lab @ SUNY Binghamton under the supervision of Professor Ping Yang and Professor Guanhua Yan where my current research focuses on improving the intersection between Machine Learning (ML) and Cybersecurity. From a ML perspective I work on adversarial machine learning where we ask the questions “are ML models used for anomaly detection robust? can they be fooled?”, I also work on interpretable machine learning where we ask the questions “Why/How do ML models for anomaly detection really work?”

In my PhD, so far I have worked on developing a model with real-time anomaly detection capability, we also tested its ability to provide anomaly root-causes. This was followed by our recent work on understanding the adversarial robustness of Deep Learning based models used for anomaly detection from distributed system logs. Additionally, I have also worked on research projects which utilize Ethereum-Blockchain for building secure, distributed systems.

In terms of past research experience, before my PhD-I have worked on research projects spanning the areas of wireless communications and robotics. During which, I gained experience in Markovian modeling, Optimizations, Deep Learning and Robotic simulations. As of August 2020 I am also a member of the International Research Council (IRC) in Sri Lanka.

Publications

Please refer my Google scholar profile for the most up to date version.

  1. Real-Time Evasion Attacks against Deep Learning-Based Anomaly Detection from Distributed System Logs. By J. Dinal Herath, Ping Yang and Guanhua Yan. In: Proceedings of The 11th ACM Conference on Data and Application Security and Privacy (CODASPY) (2021). [Accepted]

  2. RAMP: Real-Time Anomaly Detection in Scientific Workflows. By J. Dinal Herath, Changxin Bai, Guanhua Yan, Ping Yang, Shiyong Lu. In: IEEE International Conference on Big Data (2019). [extended version] [code][slides]

  3. SciBlock: A Blockchain-Based Tamper-Proof Non-Repudiable Storage for Scientific Workflow Provenance. By Dinuni Fernando, Siddharth Kulshrestha, J. Dinal Herath, Nitin Mahadik, Yanzhe Ma, Changxin Bai, Ping Yang, Guanhua Yan, Shiyong Lu. In: IEEE International Conference on Collaboration and Internet Computing (2019). [slides]

  4. DeepChannel: Wireless Channel Quality Prediction using Deep Learning. By Adita Kulkarni, Anand Seetharam, Arti Ramesh, J. Dinal Herath. In: IEEE Transactions in Vehicular Technology (TVT) (2019).

  5. A Deep Learning Model for Wireless Channel Quality Prediction by J. Dinal Herath, Anand Seetharam and Arti Ramesh. In: IEEE International Conference on Communications (ICC) (2019).

  6. A Markovian Model for Analyzing Opportunistic Request Routing in Wireless Cache Networks by J. Dinal Herath and Anand Seetharam. In: IEEE Transactions in Vehicular Technology (TVT) (2018). [code] [project] [slides]

  7. Analyzing Opportunistic Request Routing in Wireless Cache Networks by J. Dinal Herath and Anand Seetharam. In: IEEE International Conference on Communications (ICC) (2018). [code] [project]

  8. Simulation of Symmetric and Asymmetric movement gaits for Lateral Undulation in Serial Snake Robots by J. Dinal Herath and K. Jayananda. In: International Conference on Computational Modeling & Simulation (ICCMS) (2017). [project]

  9. Comparison of Serial and Parallel Snake Robots for Lateral Undulation Motion Using Gazebo by J. Dinal Herath and K. Jayananda. In: IEEE International Conference on Information and Automation for Sustainability (ICIAfS) (2016). (DOI - 10.1109/ICIAFS.2016.7946540) [code] [project]

Thesis