DeepAI AI Chat
Log In Sign Up

The Indian Chefs Process

by   Patrick Dallaire, et al.

This paper introduces the Indian Chefs Process (ICP), a Bayesian nonparametric prior on the joint space of infinite directed acyclic graphs (DAGs) and orders that generalizes Indian Buffet Processes. As our construction shows, the proposed distribution relies on a latent Beta Process controlling both the orders and outgoing connection probabilities of the nodes, and yields a probability distribution on sparse infinite graphs. The main advantage of the ICP over previously proposed Bayesian nonparametric priors for DAG structures is its greater flexibility. To the best of our knowledge, the ICP is the first Bayesian nonparametric model supporting every possible DAG. We demonstrate the usefulness of the ICP on learning the structure of deep generative sigmoid networks as well as convolutional neural networks.


page 1

page 7

page 8


Priors on exchangeable directed graphs

Directed graphs occur throughout statistical modeling of networks, and e...

Conjugate Projective Limits

We characterize conjugate nonparametric Bayesian models as projective li...

The Infinite Latent Events Model

We present the Infinite Latent Events Model, a nonparametric hierarchica...

Deep Bayesian Nonparametric Factor Analysis

We propose a deep generative factor analysis model with beta process pri...

A Prior Distribution over Directed Acyclic Graphs for Sparse Bayesian Networks

The main contribution of this article is a new prior distribution over d...

Nonparametric Bayesian estimation of multivariate Hawkes processes

This paper studies nonparametric estimation of parameters of multivariat...

Learning Sparse Nonparametric DAGs

We develop a framework for learning sparse nonparametric directed acycli...