Creighton Heaukulani

I am a machine learning researcher at Goldman Sachs in Hong Kong.

We are always looking to hire talented machine learning practitioners into any of our major offices. Feel free to contact me for more information if you are interested:

c [dot] k [dot] heaukulani [at]

My research interests are in probabilistic approaches to modeling and inference in all areas of modern statistics.

I completed my PhD in the Machine Learning Group at the University of Cambridge, under the supervision of Zoubin Ghahramani. I was previously supervised by Andrew R. Barron and Sekhar Tatikonda in the Statistics Department at Yale University.


Bayesian inference on random simple graphs with power law degree distributions.
Juho Lee, C. Heaukulani, Zoubin Ghahramani, Lancelot F. James, and Seungjin Choi
ICML, 2017. [paper]

Black-box constructions for exchangeable sequences of random multisets.
C. Heaukulani and Daniel M. Roy
Preprint available upon request.

Generalized IBPs, random multisets, and tree-structured feature allocations.
PhD Thesis.
University of Cambridge, Sept. 2016. [pdf]

Gibbs-type Indian buffet processes.
C. Heaukulani and Daniel M. Roy
Preprint. [paper]

Beta diffusion trees and hierarchical feature allocations.
C. Heaukulani, David A. Knowles, and Zoubin Ghahramani
ICML, 2014. Extended version: [paper][slides]

The combinatorial structure of beta negative binomial processes.
C. Heaukulani and Daniel M. Roy
Bernoulli, Vol. 22, No. 4, 2301--2324, 2016. [paper][slides]

Dynamic probabilistic models for latent feature propagation in social networks.
C. Heaukulani and Zoubin Ghahramani
ICML, 2013. [paper][slides]


Random partition based inference schemes for feature allocations.
BNP 10, Raleigh--Durham, June 2015 [slides]

Beta diffusion trees.
ICML 2014, Beijing, June 2014 [slides]

Probabilistic latent feature propagation in social networks.
ICML 2013, Atlanta, June 2013 [slides]
NetSci 2013 Satellite Symposium, Copenhagen, June 2013

The negative binomial IBP.
BNP 9, Amsterdam, March 2013 [slides]