Browse Source

Add model capabilities table

arxiv-v3
Markus Kaiser 3 years ago
parent
commit
5240e58cb4
  1. 34
      dynamic_dirichlet_deep_gp.tex

34
dynamic_dirichlet_deep_gp.tex

@ -352,6 +352,40 @@ This extended bound thus has complexity $\Fun*{\Oh}{NM^2LK}$ to evaluate in the
\section{Experiments}
\label{sec:experiments}
\begin{table}[t]
\centering
\caption{
\label{tab:model_capabilities}
Model Capabilities
}
\scriptsize
\newcolumntype{Y}{>{\centering\arraybackslash}X}%
\newcommand{\yes}{\checkmark}
\newcommand{\no}{--}
\newcommand{\resultrow}[9]{#1 & #4 & #7 & #3 & #9 & #5 & #6 & #8 \\}
% \setlength{\tabcolsep}{1pt}
\begin{tabularx}{\linewidth}{lYYYYYYYY}
\toprule
% NOTE(mrksr): Original order
% Model & Bayesian & Scalable Inference & Approximate Joint & Data Association & Predictive Associations & Multimodal Data & Per-Mode Predictions & Interpretable Priors \\
% NOTE(mrksr): Presentation order
& Joint Predictions & Multimodal Data & Scalable Inference & Interpretable Priors & Data Associations & Predictive Associations & Per-Process Predictions \\
\midrule
Experiment & & & & & \cref{subsec:semi_bimodal} & \cref{subsec:choicenet} & \cref{subsec:cartpole} \\
\midrule
\resultrow{DAGP}{\yes}{\yes}{\yes}{\yes}{\yes}{\yes}{\yes}{\yes}
\addlinespace
\resultrow{RGPR \parencite{lazaro-gredilla_overlapping_2012}}{\yes}{\no}{\yes}{\yes}{\no}{\yes}{\no}{\yes}
\resultrow{MLE \parencite{tresp_mixtures_2001}}{\yes}{\no}{\yes}{\no}{\no}{\no}{\no}{\yes}
\resultrow{LatentGP \parencite{bodin_latent_2017}}{\yes}{\no}{\yes}{\no}{\no}{\yes}{\no}{\yes}
\resultrow{GPR}{\yes}{\yes}{\yes}{\no}{\no}{\no}{\no}{\yes}
\addlinespace
\resultrow{MLP}{\no}{\yes}{\yes}{\no}{\no}{\no}{\no}{\no}
\resultrow{MDN \parencite{bishop_mixture_1994}}{\no}{\yes}{\yes}{\no}{\no}{\yes}{\no}{\no}
\resultrow{BNN+LV \parencite{depeweg_learning_2016}}{\yes}{\yes}{\yes}{\no}{\no}{\yes}{\no}{\no}
\bottomrule
\end{tabularx}
\end{table}
In this section we investigate the behavior of the DAGP model in multiple regression settings.
First, we show how prior knowledge about the different generative processes can be used to separate a signal from unrelated noise.
Second, we apply the DAGP to a multimodal data set and showcase how the different components of the model interact to identify how many modes are necessary to explain the data.

Loading…
Cancel
Save