S
Svetha Venkatesh
Researcher at Deakin University
Publications - 864
Citations - 20118
Svetha Venkatesh is an academic researcher from Deakin University. The author has contributed to research in topics: Bayesian optimization & Computer science. The author has an hindex of 60, co-authored 828 publications receiving 16441 citations. Previous affiliations of Svetha Venkatesh include Australian National University & National University of Singapore.
Papers
More filters
An extended frequent pattern tree for intertransaction association rule mining
TL;DR: Experimental results show significant computational improvement of the EFP-Tree over FITI when a large number of rules is present in the data.
Proceedings Article
A nonparametric Bayesian Poisson gamma model for count data
TL;DR: This work proposes a nonparametric Bayesian, linear Poisson gamma model for count data and uses it for dictionary learning and presents an auxiliary variable Gibbs sampler, which turns the intractable inference into a tractable one.
Proceedings ArticleDOI
Probabilistic models over ordered partitions with applications in document ranking and collaborative filtering
Dinh Phung,Svetha Venkatesh +1 more
TL;DR: A probabilistic generative model is proposed, that models the process of ranking documents in super-exponential combinatorial state space with unknown numbers of partitions and unknown ordering among them, and it is shown that with suitable parameterisation, the models can be learned in linear time.
Journal ArticleDOI
Solving for X: Evidence for sex‐specific autism biomarkers across multiple transcriptomic studies
Samuel C. Lee,Thomas P. Quinn,Jerry Lai,Sek Won Kong,Sek Won Kong,Irva Hertz-Picciotto,Stephen J. Glatt,Tamsyn M. Crowley,Tamsyn M. Crowley,Svetha Venkatesh,Thin Nguyen +10 more
TL;DR: Through pathway analysis, it is found that these sex‐independent biomarkers have substantially different biological roles than the sex‐dependent biomarkers, and that some of these pathways are ubiquitously dysregulated in both postmortem brain and blood.
Journal ArticleDOI
Energy-based anomaly detection for mixed data
TL;DR: The results demonstrate that for anomaly detection, a proper handling of mixed types is necessary, free energy is a powerful anomaly scoring method, multilevel abstraction of data is important for high-dimensional data, and empirically Mv.RBM and$$MIXMAD are superior to popular unsupervised detection methods for both homogeneous and mixed data.