Dirichlet Process

By Yee Whye Teh. 2017

353 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Mathematical Physics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Mathematical Physics

Citations in the corpus

A Neural Dirichlet Process Mixture Model for Task-Free Continual Learning (2020, background): “For a distribution G sampled from DP(α,G0), the following holds for any finite measurable partition {A1, A2, ..., AK} of probability space Θ (Teh, 2010): (G(A1), ..., G(AK)) ∼ Dir(αG0(A1), ..., αG0(AK)).”
Continual Unsupervised Representation Learning (2019, methodology): “In a similar fashion to existing techniques such as the Forget-Me-Not process (Milan et al., 2016) and Dirichlet process (Teh, 2010), we rely on a threshold to determine when to instantiate a new component.”
Emergence of Grounded Compositional Language in Multi-Agent Populations (2017, methodology): “Inspired by these rich-get-richer dynamics, we model the communication symbols as being generated from a Dirichlet Process (Teh 2011).”
Recent Advances in Open Set Recognition: A Survey (2018, neutral): “2) Non-Instance Generation-based OSR Models: Dirichlet process (DP) [119]–[123] considered as a distribution over distributions is a stochastic process, which has been widely applied in clustering and density estimation problems as a nonparametric prior defined over the number of mixture components.”
Been There, Done That: Meta-Learning with Episodic Recall (2018, background): “Otherwise, it draws a task from the urn uniformly at random, copies it, and drops both the original and new copy into the urn (Teh, 2011).”
Optimism-driven exploration for nonlinear systems (2015, methodology): “mixture model (DPMM) [13] for the dynamics, which strikes a good balance between expressiveness and computational complexity, as discussed in Section V.”
Learning Object Arrangements in 3D Scenes using Human Context (2012, methodology): “The inference in our model is similar to the standard DP (see Teh (2010) and Neal (2000) for an overview).”
Ontology-Based Semantic Image Segmentation Using Mixture Models and Multiple CRFs (2016, methodology): “The Dirichlet process mixture model (infinite mixture model) used in Bayesian nonparametrics modeling is a distribution over distributions [37].”