Alert button
Picture for David Duvenaud

David Duvenaud

Alert button

Explaining Image Classifiers by Counterfactual Generation

Oct 11, 2018
Chun-Hao Chang, Elliot Creager, Anna Goldenberg, David Duvenaud

Figure 1 for Explaining Image Classifiers by Counterfactual Generation
Figure 2 for Explaining Image Classifiers by Counterfactual Generation
Figure 3 for Explaining Image Classifiers by Counterfactual Generation
Figure 4 for Explaining Image Classifiers by Counterfactual Generation
Viaarxiv icon

Stochastic Combinatorial Ensembles for Defending Against Adversarial Examples

Sep 08, 2018
George A. Adam, Petr Smirnov, David Duvenaud, Benjamin Haibe-Kains, Anna Goldenberg

Figure 1 for Stochastic Combinatorial Ensembles for Defending Against Adversarial Examples
Figure 2 for Stochastic Combinatorial Ensembles for Defending Against Adversarial Examples
Figure 3 for Stochastic Combinatorial Ensembles for Defending Against Adversarial Examples
Figure 4 for Stochastic Combinatorial Ensembles for Defending Against Adversarial Examples
Viaarxiv icon

Scalable Recommender Systems through Recursive Evidence Chains

Jul 05, 2018
Elias Tragas, Calvin Luo, Maxime Gazeau, Kevin Luk, David Duvenaud

Figure 1 for Scalable Recommender Systems through Recursive Evidence Chains
Figure 2 for Scalable Recommender Systems through Recursive Evidence Chains
Viaarxiv icon

Inference Suboptimality in Variational Autoencoders

May 27, 2018
Chris Cremer, Xuechen Li, David Duvenaud

Figure 1 for Inference Suboptimality in Variational Autoencoders
Figure 2 for Inference Suboptimality in Variational Autoencoders
Figure 3 for Inference Suboptimality in Variational Autoencoders
Figure 4 for Inference Suboptimality in Variational Autoencoders
Viaarxiv icon

Stochastic Hyperparameter Optimization through Hypernetworks

Mar 08, 2018
Jonathan Lorraine, David Duvenaud

Figure 1 for Stochastic Hyperparameter Optimization through Hypernetworks
Figure 2 for Stochastic Hyperparameter Optimization through Hypernetworks
Figure 3 for Stochastic Hyperparameter Optimization through Hypernetworks
Figure 4 for Stochastic Hyperparameter Optimization through Hypernetworks
Viaarxiv icon

Noisy Natural Gradient as Variational Inference

Feb 26, 2018
Guodong Zhang, Shengyang Sun, David Duvenaud, Roger Grosse

Figure 1 for Noisy Natural Gradient as Variational Inference
Figure 2 for Noisy Natural Gradient as Variational Inference
Figure 3 for Noisy Natural Gradient as Variational Inference
Figure 4 for Noisy Natural Gradient as Variational Inference
Viaarxiv icon

Backpropagation through the Void: Optimizing control variates for black-box gradient estimation

Feb 23, 2018
Will Grathwohl, Dami Choi, Yuhuai Wu, Geoffrey Roeder, David Duvenaud

Figure 1 for Backpropagation through the Void: Optimizing control variates for black-box gradient estimation
Figure 2 for Backpropagation through the Void: Optimizing control variates for black-box gradient estimation
Figure 3 for Backpropagation through the Void: Optimizing control variates for black-box gradient estimation
Figure 4 for Backpropagation through the Void: Optimizing control variates for black-box gradient estimation
Viaarxiv icon

Generating and designing DNA with deep generative models

Dec 17, 2017
Nathan Killoran, Leo J. Lee, Andrew Delong, David Duvenaud, Brendan J. Frey

Figure 1 for Generating and designing DNA with deep generative models
Figure 2 for Generating and designing DNA with deep generative models
Figure 3 for Generating and designing DNA with deep generative models
Figure 4 for Generating and designing DNA with deep generative models
Viaarxiv icon

Automatic chemical design using a data-driven continuous representation of molecules

Dec 05, 2017
Rafael Gómez-Bombarelli, Jennifer N. Wei, David Duvenaud, José Miguel Hernández-Lobato, Benjamín Sánchez-Lengeling, Dennis Sheberla, Jorge Aguilera-Iparraguirre, Timothy D. Hirzel, Ryan P. Adams, Alán Aspuru-Guzik

Figure 1 for Automatic chemical design using a data-driven continuous representation of molecules
Figure 2 for Automatic chemical design using a data-driven continuous representation of molecules
Figure 3 for Automatic chemical design using a data-driven continuous representation of molecules
Figure 4 for Automatic chemical design using a data-driven continuous representation of molecules
Viaarxiv icon

Reinterpreting Importance-Weighted Autoencoders

Aug 15, 2017
Chris Cremer, Quaid Morris, David Duvenaud

Figure 1 for Reinterpreting Importance-Weighted Autoencoders
Figure 2 for Reinterpreting Importance-Weighted Autoencoders
Figure 3 for Reinterpreting Importance-Weighted Autoencoders
Viaarxiv icon