Michael Houle

I received my PhD degree in 1989 from McGill University in Canada, in the area of computational geometry, and my research interests have been shifting ever since. After graduation, I spent almost 3 years in Japan, and then 9 years in Australia at the University of Newcastle and the University of Sydney, during which time it appeared that I had found my niche in algorithmics and relational visualization. In 2001 I was back in Japan, where at IBM's Tokyo Research Laboratory my main focus was on approximate similarity search and shared-neighbor clustering methods for data mining applications. From 2004 to 2021, at the National Institute of Informatics, Tokyo, my theoretical interests were concentrated on dimensionality and scalability, while targeting applications in the context of fundamental AI-ML-DM tasks such as search, clustering, classification, and outlier detection. During the pandemic, I returned to Canada, setting up home in the Vancouver area where I conduct my teaching and research for NJIT, with occasional visits to the main campus.

Ph.D.; McGill University; Computer Science; 1989

B.Sc.; McGill University; Mathematics and Computer Science; 1984

2024 Best Research Paper Award, SIAM International Conference on Data Mining (SDM 2024)

2021 Best Paper Award, 14th International Conference on Similarity Search and Applications (SISAP 2021)

2018 Best Paper Award, 8th International Conference on Web Intelligence, Mining and Semantics (WIMS 2018)

2014 Best Paper Award, 7th International Conference on Similarity Search and Applications (SISAP 2014)

2010 Best Research Paper Award, 10th IEEE International Conference on Data Mining (ICDM 2010)

https://scholar.google.com/citations?user=24wEYXoAAAAJ

DS 675 - MACHINE LEARNING

Machine Learning, Data Mining, Data Structures & Algorithms

Machine Learning, Data Mining, Similarity Search, Algorithmics, Extreme Value Theory

Dimensionality-Aware Outlier Detection
2024 SIAM International Conference on Data Mining (SDM), April (2nd Quarter/Spring) 2024

LDReg: Local Dimensionality Regularized Self-Supervised Learning
2024 International Conference on Learning Representations (ICLR), February 2024

Relationships Between Local Intrinsic Dimensionality and Tail Entropy
2021 International Conference on Similarity Search and Applications (SISAP), October (4th Quarter/Autumn) 2021

The Effect of Random Projection on Local Intrinsic Dimensionality
2021 International Conference on Similarity Search and Applications (SISAP), October (4th Quarter/Autumn) 2021

A Dimensionality-Driven Approach for Unsupervised Out-of-distribution Detection
2021 SIAM International Conference on Data Mining (SDM), April (2nd Quarter/Spring) 2021

Bailey, James, & Houle, Michael E. , & Ma, Xingjun (2023). Relationships between Tail Entropies and Local Intrinsic Dimensionality and their Use for Estimation and Feature Representation. Information Systems, 118(102245),

Bailey, James, & Houle, Michael E., & Ma, Xingjun (2022). Local Intrinsic Dimensionality, Entropy and Statistical Divergences. Entropy, 24(9), 1220.

Amsaleg, Laurent, & Bailey, James, & Barbe, Amélie, & Erfani, Sarah M., & Furon, Teddy, & Houle, Michael E., & Radovanović, Miloš, & Nguyen, Xuan Vinh (2021). High Intrinsic Dimensionality Facilitates Adversarial Attack: Theoretical Evidence. IEEE Transactions on Information Forensics and Security (TIFS), 16, 854 - 865.

Houle, Michael E., & Kiermeier, Marie, & Zimek, Arthur (2023). Clustering High-Dimensional Data, Lior Rokach, Oded Maimon, Erez Shmueli (Eds.), Machine Learning for Data Science Handbook: Data Mining and Knowledge Discovery Handbook (Springer). (pp. 219-237). Machine Learning for Data Science Handbook: Data Mining and Knowledge Discovery Handbook (Springer)