4 on Diana Cai https://dicai.github.io/publication_types/4/ Recent content in 4 on Diana Cai Hugo -- gohugo.io en-us Wed, 02 Nov 2022 00:00:00 +0000 Multi-fidelity Bayesian experimental design using power posteriors https://dicai.github.io/publication/conference-paper/2022-multi-fidelity-experimental-design/ Wed, 02 Nov 2022 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2022-multi-fidelity-experimental-design/ Efficient Bayesian inverse reinforcement learning via conditional kernel density estimation https://dicai.github.io/publication/conference-paper/2022-birl-aabi/ Thu, 27 Oct 2022 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2022-birl-aabi/ Slice sampling reparameterization gradients https://dicai.github.io/publication/conference-paper/2021-reparam-slice-wkshp/ Fri, 01 Jan 2021 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2021-reparam-slice-wkshp/ Power posteriors do not reliably learn the number of components in a finite mixture https://dicai.github.io/publication/conference-paper/2020-power-mix/ Thu, 08 Oct 2020 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2020-power-mix/ See also: Finite mixture models do not reliably learn the number of components (arXiv e-print 2007.04470) for results on learning the number of components in a finite mixture model via the usual posterior distribution. Weighted meta-learning https://dicai.github.io/publication/conference-paper/2020-weighted-meta-learning-wkshp/ Wed, 01 Jul 2020 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2020-weighted-meta-learning-wkshp/ A longer version of this article is available on the arXiv (link). Finite mixture models are typically inconsistent for the number of components https://dicai.github.io/publication/conference-paper/2019-finmix/ Sun, 01 Dec 2019 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2019-finmix/ Finite mixture models are typically inconsistent for the number of components https://dicai.github.io/publication/conference-paper/2017-finmix/ Fri, 01 Dec 2017 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2017-finmix/ Paintboxes and probability functions for edge-exchangeable graphs https://dicai.github.io/publication/conference-paper/2016-paintbox/ Thu, 01 Dec 2016 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2016-paintbox/ In NeurIPS 2016 Workshop on Adaptive and Scalable Nonparametric Methods in Machine Learning Completely random measures for modeling power laws in graphs https://dicai.github.io/publication/conference-paper/2015-powerlaws/ Tue, 01 Dec 2015 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2015-powerlaws/ Edge-exchangeable graphs and sparsity https://dicai.github.io/publication/conference-paper/2015-edgeexch-networks/ Tue, 01 Dec 2015 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2015-edgeexch-networks/ In NeurIPS 2015 Workshop on Networks in the Social and Information Sciences Edge-exchangeable graphs, sparsity, and power laws https://dicai.github.io/publication/conference-paper/2015-edgeexch-bnp/ Tue, 01 Dec 2015 00:00:00 +0000 https://dicai.github.io/publication/conference-paper/2015-edgeexch-bnp/ In NeurIPS 2015 Workshop on Bayesian Nonparametrics