Picture for Emily Fertig

Emily Fertig

Embedded-model flows: Combining the inductive biases of model-free deep learning and explicit probabilistic modeling

Add code
Oct 17, 2021
Figure 1 for Embedded-model flows: Combining the inductive biases of model-free deep learning and explicit probabilistic modeling
Figure 2 for Embedded-model flows: Combining the inductive biases of model-free deep learning and explicit probabilistic modeling
Figure 3 for Embedded-model flows: Combining the inductive biases of model-free deep learning and explicit probabilistic modeling
Figure 4 for Embedded-model flows: Combining the inductive biases of model-free deep learning and explicit probabilistic modeling
Viaarxiv icon

Likelihood Ratios for Out-of-Distribution Detection

Add code
Jun 07, 2019
Figure 1 for Likelihood Ratios for Out-of-Distribution Detection
Figure 2 for Likelihood Ratios for Out-of-Distribution Detection
Figure 3 for Likelihood Ratios for Out-of-Distribution Detection
Figure 4 for Likelihood Ratios for Out-of-Distribution Detection
Viaarxiv icon

Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift

Add code
Jun 06, 2019
Figure 1 for Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
Figure 2 for Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
Figure 3 for Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
Figure 4 for Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
Viaarxiv icon

Dueling Decoders: Regularizing Variational Autoencoder Latent Spaces

May 17, 2019
Figure 1 for Dueling Decoders: Regularizing Variational Autoencoder Latent Spaces
Figure 2 for Dueling Decoders: Regularizing Variational Autoencoder Latent Spaces
Figure 3 for Dueling Decoders: Regularizing Variational Autoencoder Latent Spaces
Figure 4 for Dueling Decoders: Regularizing Variational Autoencoder Latent Spaces
Viaarxiv icon

$β$-VAEs can retain label information even at high compression

Dec 06, 2018
Figure 1 for $β$-VAEs can retain label information even at high compression
Figure 2 for $β$-VAEs can retain label information even at high compression
Viaarxiv icon