Published in Transactions on Machine Learning Research (TMLR), 2023
PrivGnn combines distillation and noise mechanisms to release graph models with rigorous privacy guarantees.
Recommended citation: Iyiola E. Olatunji, Thorben Funke, and Megha Khosla. (2023). "Releasing graph neural networks with differential privacy guarantees." Transactions on Machine Learning Research (TMLR).
Download Paper
Published in arXiv, 2023
We investigate whether black-box attribute inference on GNNs creates meaningful privacy risk.
Recommended citation: Iyiola E. Olatunji, Anmar Hizber, Oliver Sihlovec, and Megha Khosla. (2023). "Does Black-box Attribute Inference Attacks on Graph Neural Networks Constitute Privacy Risk?" arXiv.
Download Paper
Published in Big Data, 2022
A review of anonymization methods, attacks, and practical privacy tools for healthcare data.
Recommended citation: Iyiola E. Olatunji, Jens Rauch, Matthias Katzensteiner, and Megha Khosla. (2022). "A review of anonymization for healthcare data." Big Data.
Download Paper
Published in arXiv, 2021
We introduce a differentially private k-NN outlier detection method based on data partitioning.
Recommended citation: Jens Rauch, Iyiola E. Olatunji, and Megha Khosla. (2021). "Achieving differential privacy for k-nearest neighbors based outlier detection by data partitioning." arXiv.
Download Paper
Published in Machine Learning Paradigms: Applications of Learning and Analytics in Intelligent Systems, 2019
A survey of video analytics methods and their applications across surveillance and related domains.
Recommended citation: Iyiola E. Olatunji and Chun-Hung Cheng. (2019). "Video analytics for visual surveillance and applications: An overview and survey." Machine Learning Paradigms: Applications of Learning and Analytics in Intelligent Systems.
Download Paper
Published in Proceedings on Privacy Enhancing Technologies (PETS), 2023
We show how model explanations can leak graph structure and study defenses against that leakage.
Recommended citation: Iyiola E. Olatunji, Mandeep Rathee, Thorben Funke, and Megha Khosla. (2023). "Private graph extraction via feature explanations." Proceedings on Privacy Enhancing Technologies (PETS).
Download Paper
Published in 2021 Third IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA), 2021
We study how graph neural networks leak membership information and propose effective defenses.
Recommended citation: Iyiola E. Olatunji, Wolfgang Nejdl, and Megha Khosla. (2021). "Membership inference attack on graph neural networks." 2021 Third IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA).
Download Paper
Published in Asia Information Retrieval Symposium, 2019
A neural model for review helpfulness prediction using context-aware text encoding.
Recommended citation: Iyiola E. Olatunji, Xin Li, and Wai Lam. (2019). "Context-Aware Helpfulness Prediction for Online Product Reviews." Asia Information Retrieval Symposium.
Download Paper
Published in Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support (DLMIA 2018), 2018
A multi-task deep learning framework for lung nodule classification and attribute scoring.
Recommended citation: Lihao Liu, Qi Dou, Hao Chen, Iyiola E. Olatunji, Jing Qin, and Pheng-Ann Heng. (2018). "Mtmr-net: Multi-task deep learning with margin ranking loss for lung nodule analysis." Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support (DLMIA 2018).
Download Paper