Generative Models for Learning from Crowds

06/13/2017

∙

In this paper, we propose generative probabilistic models for label aggregation. We use Gibbs sampling and a novel variational inference algorithm to perform the posterior inference. Empirical results show that our methods consistently outperform state-of-the-art methods.

READ FULL TEXT

Generative Models for Learning from Crowds

Amortized Inference of Variational Bounds for Learning Noisy-OR

Amortized Population Gibbs Samplers with Neural Sufficient Statistics

Variational Measure Preserving Flows

Hidden Tree Markov Networks: Deep and Wide Learning for Structured Data

Model-based micro-data reinforcement learning: what are the crucial model properties and which model to choose?

Faithful Model Inversion Substantially Improves Auto-encoding Variational Inference

Plug-and-Play split Gibbs sampler: embedding deep generative priors in Bayesian inference

Generative Models for Learning from Crowds

Related Research

Amortized Inference of Variational Bounds for Learning Noisy-OR

Amortized Population Gibbs Samplers with Neural Sufficient Statistics

Variational Measure Preserving Flows

Hidden Tree Markov Networks: Deep and Wide Learning for Structured Data

Model-based micro-data reinforcement learning: what are the crucial model properties and which model to choose?

Faithful Model Inversion Substantially Improves Auto-encoding Variational Inference

Plug-and-Play split Gibbs sampler: embedding deep generative priors in Bayesian inference