Compare commits
2 commits
9a564750e2
...
61b5f3ccbd
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
61b5f3ccbd | ||
|
|
feb2b46a0a |
3 changed files with 184 additions and 262 deletions
99
.obsidian/workspace.json
vendored
99
.obsidian/workspace.json
vendored
|
|
@ -6,63 +6,20 @@
|
||||||
{
|
{
|
||||||
"id": "1cf21b6cf48676e1",
|
"id": "1cf21b6cf48676e1",
|
||||||
"type": "tabs",
|
"type": "tabs",
|
||||||
"dimension": 55.52282768777614,
|
|
||||||
"children": [
|
|
||||||
{
|
|
||||||
"id": "9008c2c9f7c08ed0",
|
|
||||||
"type": "leaf",
|
|
||||||
"pinned": true,
|
|
||||||
"state": {
|
|
||||||
"type": "markdown",
|
|
||||||
"state": {
|
|
||||||
"file": "Thèse/Articles/Review papier colBiSBM.md",
|
|
||||||
"mode": "source",
|
|
||||||
"source": false,
|
|
||||||
"backlinks": false
|
|
||||||
},
|
|
||||||
"pinned": true,
|
|
||||||
"icon": "lucide-file",
|
|
||||||
"title": "Review papier colBiSBM"
|
|
||||||
}
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"id": "bd694dfa434e6a04",
|
|
||||||
"type": "leaf",
|
|
||||||
"state": {
|
|
||||||
"type": "markdown",
|
|
||||||
"state": {
|
|
||||||
"file": "Thèse/Packages/R/colSBM.md",
|
|
||||||
"mode": "source",
|
|
||||||
"source": false,
|
|
||||||
"backlinks": false
|
|
||||||
},
|
|
||||||
"icon": "lucide-file",
|
|
||||||
"title": "colSBM"
|
|
||||||
}
|
|
||||||
}
|
|
||||||
],
|
|
||||||
"currentTab": 1
|
|
||||||
},
|
|
||||||
{
|
|
||||||
"id": "093b69f5907f4d4c",
|
|
||||||
"type": "tabs",
|
|
||||||
"dimension": 44.47717231222386,
|
|
||||||
"children": [
|
"children": [
|
||||||
{
|
{
|
||||||
"id": "5a22d122169dedf7",
|
"id": "5a22d122169dedf7",
|
||||||
"type": "leaf",
|
"type": "leaf",
|
||||||
"pinned": true,
|
|
||||||
"state": {
|
"state": {
|
||||||
"type": "markdown",
|
"type": "markdown",
|
||||||
"state": {
|
"state": {
|
||||||
"file": "Résumé des tâches.md",
|
"file": "Thèse/Axes/Phylogénie/SBM avec covariance latente.md",
|
||||||
"mode": "source",
|
"mode": "source",
|
||||||
"source": false,
|
"source": false,
|
||||||
"backlinks": false
|
"backlinks": false
|
||||||
},
|
},
|
||||||
"pinned": true,
|
|
||||||
"icon": "lucide-file",
|
"icon": "lucide-file",
|
||||||
"title": "Résumé des tâches"
|
"title": "SBM avec covariance latente"
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
]
|
]
|
||||||
|
|
@ -240,9 +197,19 @@
|
||||||
"icon": "quote-glyph",
|
"icon": "quote-glyph",
|
||||||
"title": "References"
|
"title": "References"
|
||||||
}
|
}
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"id": "6734703bbf1b5f03",
|
||||||
|
"type": "leaf",
|
||||||
|
"state": {
|
||||||
|
"type": "git-view",
|
||||||
|
"state": {},
|
||||||
|
"icon": "git-pull-request",
|
||||||
|
"title": "Source Control"
|
||||||
|
}
|
||||||
}
|
}
|
||||||
],
|
],
|
||||||
"currentTab": 7
|
"currentTab": 8
|
||||||
}
|
}
|
||||||
],
|
],
|
||||||
"direction": "horizontal",
|
"direction": "horizontal",
|
||||||
|
|
@ -268,43 +235,43 @@
|
||||||
"templater-obsidian:Templater": false
|
"templater-obsidian:Templater": false
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
"active": "bd694dfa434e6a04",
|
"active": "5a22d122169dedf7",
|
||||||
"lastOpenFiles": [
|
"lastOpenFiles": [
|
||||||
|
"Thèse/Axes/Phylogénie/SBM avec covariance multinomiale probit.md",
|
||||||
|
"Thèse/Axes/Phylogénie/SBM avec covariance latente.md",
|
||||||
"Résumé des tâches.md",
|
"Résumé des tâches.md",
|
||||||
"Thèse/Résolution des problèmes/Problème avec renv.md",
|
"Thèse/StateOfTheR/HappyR/2026-05-29 MLOps en Python.md",
|
||||||
"Thèse/Packages/R/colSBM.md",
|
"Thèse/Packages/R/colSBM.md",
|
||||||
"Thèse/Articles/Review papier colBiSBM.md",
|
"Thèse/Projets annexes/VGAE avec (Gromov-)Wasserstein.md",
|
||||||
|
"Thèse/Projets annexes/Applications colBiSBM pour impact pratiques agris sur interactions plantes pollinisateurs.md",
|
||||||
|
"Thèse/Projets annexes/Application colBiSBM réseaux d'optimisation de NN.md",
|
||||||
|
"Thèse/Projets annexes",
|
||||||
"Thèse/Packages/R",
|
"Thèse/Packages/R",
|
||||||
"Thèse/Packages",
|
"Thèse/Packages",
|
||||||
"Thèse/Projets annexes/Application colBiSBM réseaux d'optimisation de NN.md",
|
"public/index-listing.json",
|
||||||
"Thèse/Projets annexes/Applications colBiSBM pour impact pratiques agris sur interactions plantes pollinisateurs.md",
|
"public/Thèse/Utile/Doranum.html",
|
||||||
"Thèse/TODO/2026-05-18.md",
|
"public/Thèse/Utile",
|
||||||
"macros.tex.md",
|
"public/Thèse/TODO/2026-05-18.html",
|
||||||
"Thèse/Lectures/local_macros.tex.md",
|
"public/Thèse/TODO",
|
||||||
"Thèse/Projets annexes/VGAE avec (Gromov-)Wasserstein.md",
|
"public/Thèse/Séminaires/2026-06-09 FRANCHETERRE Blanche.html",
|
||||||
"Thèse/Projets annexes",
|
"public/Thèse/Séminaires/2026-05-21 CHENNETIER Guillaume.html",
|
||||||
"Thèse/Axes/Phylogénie/SBM avec covariance latente.md",
|
|
||||||
"Thèse/Lectures/@quericBridgingMaximumLikelihood2026.md",
|
"Thèse/Lectures/@quericBridgingMaximumLikelihood2026.md",
|
||||||
"@quericBridgingMaximumLikelihood2026.md",
|
"@quericBridgingMaximumLikelihood2026.md",
|
||||||
|
"Thèse/Articles/Review papier colBiSBM.md",
|
||||||
|
"Thèse/TODO/2026-05-18.md",
|
||||||
"Thèse/Résumés séminaires.md",
|
"Thèse/Résumés séminaires.md",
|
||||||
"Thèse/Séminaires/2026-06-09 VICTOR François.md",
|
"Thèse/Séminaires/2026-06-09 VICTOR François.md",
|
||||||
"Thèse/Séminaires/2026-06-09 FRANCHETERRE Blanche.md",
|
"Thèse/Séminaires/2026-06-09 FRANCHETERRE Blanche.md",
|
||||||
"Perso/Tâches/2026-05-17.md",
|
"Perso/Tâches/2026-05-17.md",
|
||||||
"Thèse/Axes/projets-phylo.qmd",
|
|
||||||
"Thèse/Tutoriel ABC.md",
|
"Thèse/Tutoriel ABC.md",
|
||||||
"Bienvenue.md",
|
"Bienvenue.md",
|
||||||
"Thèse/Lectures/@turnerTutorialApproximateBayesian2012.md",
|
"Thèse/Lectures/@turnerTutorialApproximateBayesian2012.md",
|
||||||
"Thèse/Lectures/@hronImputationMissingValues2010.md",
|
"Thèse/Lectures/@hronImputationMissingValues2010.md",
|
||||||
"Thèse/Axes/Phylogénie/SBM avec covariance multinomiale probit.md",
|
|
||||||
"Thèse/Axes/Phylogénie/Phylogénie Papiers à regarder.md",
|
"Thèse/Axes/Phylogénie/Phylogénie Papiers à regarder.md",
|
||||||
"Perso/Tâches/Casque antibruit pour l'anniversaire de Kris.md",
|
"Perso/Tâches/Casque antibruit pour l'anniversaire de Kris.md",
|
||||||
"Thèse/StateOfTheR/HappyR/2026-05-29 MLOps en Python.md",
|
|
||||||
"@boraCompressedSensingUsing2017.md",
|
"@boraCompressedSensingUsing2017.md",
|
||||||
"Thèse/StateOfTheR/HappyR/Sans titre",
|
"Thèse/Groupes de travail/Modèles génératifs/VAE pour la régularisation de problème inverse.md",
|
||||||
"Thèse/StateOfTheR/HappyR",
|
"Thèse/Séminaires/2026-05-28 SILVA BERNARDES Juliana.md",
|
||||||
"Thèse/StateOfTheR",
|
"Meta/Modèles/Séminaire_Template.md"
|
||||||
"Thèse/Groupes de travail/Modèles génératifs",
|
|
||||||
"Thèse/Groupes de travail",
|
|
||||||
"public/index-listing.json"
|
|
||||||
]
|
]
|
||||||
}
|
}
|
||||||
|
|
@ -1,4 +1,6 @@
|
||||||
|
# Inclusion dans la thèse
|
||||||
|
|
||||||
|
Ce modèle ci est une proposition pour remettre en cause l'hypothèse du LBM d'indépendance des $Z_{i}$ en définissant une structure de dépendance dépendantes de la phylogénie.
|
||||||
|
|
||||||
# Idée du modèle
|
# Idée du modèle
|
||||||
|
|
||||||
|
|
@ -33,238 +35,191 @@ avec $\Sigma$, la matrice de variance-covariance déterminée en fonction de l'a
|
||||||
|
|
||||||
\usepackage{amsmath,amssymb}
|
\usepackage{amsmath,amssymb}
|
||||||
|
|
||||||
\usetikzlibrary{arrows.meta,positioning,shapes.geometric,calc}
|
\usetikzlibrary{
|
||||||
|
|
||||||
|
arrows.meta,
|
||||||
|
|
||||||
|
positioning,
|
||||||
|
|
||||||
|
shapes.geometric,
|
||||||
|
|
||||||
|
calc,
|
||||||
|
|
||||||
|
fit
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\begin{document}
|
\begin{document}
|
||||||
|
|
||||||
\begin{tikzpicture}[
|
\begin{tikzpicture}[
|
||||||
|
|
||||||
font=\sffamily,
|
font=\sffamily,
|
||||||
|
|
||||||
>=Latex,
|
>=Latex,
|
||||||
node distance=1.5cm and 2cm,
|
|
||||||
|
node distance=1cm and 0.25cm,
|
||||||
|
|
||||||
directed/.style={-{Latex}, line width=0.8pt, draw=gray!75},
|
directed/.style={-{Latex}, line width=0.8pt, draw=gray!75},
|
||||||
|
|
||||||
bidirected/.style={-, line width=0.8pt, draw=red!75},
|
bidirected/.style={-, line width=0.8pt, draw=red!75},
|
||||||
|
|
||||||
base/.style={
|
base/.style={
|
||||||
draw=gray!70,
|
|
||||||
line width=0.9pt,
|
draw=gray!70,
|
||||||
align=center,
|
|
||||||
inner sep=4pt
|
line width=0.9pt,
|
||||||
|
|
||||||
|
align=center,
|
||||||
|
|
||||||
|
inner sep=4pt
|
||||||
|
|
||||||
},
|
},
|
||||||
|
|
||||||
prior/.style={base,rectangle,rounded corners=1pt,fill=blue!7},
|
prior/.style={base,rectangle,rounded corners=1pt,fill=blue!7},
|
||||||
|
|
||||||
latent/.style={base,rectangle,rounded corners=6pt,fill=teal!8},
|
latent/.style={base,rectangle,rounded corners=6pt,fill=teal!8},
|
||||||
|
|
||||||
known/.style={base,diamond,aspect=1.35,fill=orange!12},
|
known/.style={base,diamond,aspect=1.35,fill=orange!12},
|
||||||
observed/.style={base,circle,fill=purple!8}
|
|
||||||
|
observed/.style={base,circle,fill=purple!8},
|
||||||
|
|
||||||
|
plate/.style={
|
||||||
|
|
||||||
|
draw=black!60,
|
||||||
|
|
||||||
|
rounded corners,
|
||||||
|
|
||||||
|
inner sep=8pt
|
||||||
|
|
||||||
|
}
|
||||||
|
|
||||||
]
|
]
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
%--------------------------------------------------
|
%--------------------------------------------------
|
||||||
% Hyperparameters rows
|
|
||||||
|
% Row clustering
|
||||||
|
|
||||||
%--------------------------------------------------
|
%--------------------------------------------------
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\node[prior] (sigma) {$\sigma^2$};
|
\node[prior] (sigma) {$\sigma^2$};
|
||||||
|
|
||||||
\node[known,right=of sigma] (Sigma) {$\Sigma$};
|
\node[known,right=of sigma] (Sigma) {$\Sigma$};
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\node[latent,below=of sigma] (Pi) {$P_i$};
|
\node[latent,below=of sigma] (Pi) {$P_i$};
|
||||||
\node[latent,below=of Sigma] (Pip) {$P_{i'}$};
|
|
||||||
|
|
||||||
\node[latent,below=of Pi] (Zi) {$Z_i$};
|
\node[latent,below=0.5cm of Pi] (Zi) {$Z_i$};
|
||||||
\node[latent,below=of Pip] (Zip) {$Z_{i'}$};
|
|
||||||
|
|
||||||
%--------------------------------------------------
|
|
||||||
% Hyperparameters columns
|
|
||||||
%--------------------------------------------------
|
|
||||||
\node[prior,right=4cm of Sigma] (rho) {$\rho$};
|
|
||||||
|
|
||||||
\node[latent,below left=of rho] (Wj) {$W_j$};
|
|
||||||
\node[latent,below right=of rho] (Wjp) {$W_{j'}$};
|
|
||||||
|
|
||||||
%--------------------------------------------------
|
|
||||||
% Intercept
|
|
||||||
%--------------------------------------------------
|
|
||||||
\node[prior] (alpha)
|
|
||||||
at ($(Zi)!0.5!(Wj)+(0,-3)$)
|
|
||||||
{$\alpha$};
|
|
||||||
|
|
||||||
%--------------------------------------------------
|
|
||||||
% Observations
|
|
||||||
%--------------------------------------------------
|
|
||||||
\node[observed]
|
|
||||||
at ($(Zi)!0.5!(Wj)+(0,-1.4)$)
|
|
||||||
(Yij) {$Y_{ij}$};
|
|
||||||
|
|
||||||
\node[observed]
|
|
||||||
at ($(Zi)!0.5!(Wjp)+(2.0,-1.4)$)
|
|
||||||
(Yijp) {$Y_{ij'}$};
|
|
||||||
|
|
||||||
\node[observed]
|
|
||||||
at ($(Zip)!0.5!(Wj)+(-2.0,-1.4)$)
|
|
||||||
(Yipj) {$Y_{i'j}$};
|
|
||||||
|
|
||||||
\node[observed]
|
|
||||||
at ($(Zip)!0.5!(Wjp)+(0,-1.4)$)
|
|
||||||
(Yipjp) {$Y_{i'j'}$};
|
|
||||||
|
|
||||||
%--------------------------------------------------
|
|
||||||
% Row side
|
|
||||||
%--------------------------------------------------
|
|
||||||
\draw[directed] (sigma) -- (Pi);
|
\draw[directed] (sigma) -- (Pi);
|
||||||
|
|
||||||
\draw[directed] (Sigma) -- (Pi);
|
\draw[directed] (Sigma) -- (Pi);
|
||||||
|
|
||||||
\draw[directed] (sigma) -- (Pip);
|
|
||||||
\draw[directed] (Sigma) -- (Pip);
|
|
||||||
|
|
||||||
\draw[directed] (Pi) -- (Zi);
|
\draw[directed] (Pi) -- (Zi);
|
||||||
\draw[directed] (Pip) -- (Zip);
|
|
||||||
|
|
||||||
|
|
||||||
%--------------------------------------------------
|
%--------------------------------------------------
|
||||||
% Column side
|
|
||||||
|
% Column clustering
|
||||||
|
|
||||||
%--------------------------------------------------
|
%--------------------------------------------------
|
||||||
|
|
||||||
|
\node[latent,right=2.5cm of Zi] (Wj) {$W_j$};
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
\node[prior] (rho) at (Wj |- sigma) {$\rho$};
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\draw[directed] (rho) -- (Wj);
|
\draw[directed] (rho) -- (Wj);
|
||||||
\draw[directed] (rho) -- (Wjp);
|
|
||||||
|
|
||||||
|
|
||||||
%--------------------------------------------------
|
%--------------------------------------------------
|
||||||
% Likelihood
|
|
||||||
%--------------------------------------------------
|
|
||||||
\foreach \y in {Yij,Yijp}
|
|
||||||
\draw[directed] (Zi) -- (\y);
|
|
||||||
|
|
||||||
\foreach \y in {Yipj,Yipjp}
|
% Observation
|
||||||
\draw[directed] (Zip) -- (\y);
|
|
||||||
|
|
||||||
\foreach \y in {Yij,Yipj}
|
|
||||||
\draw[directed] (Wj) -- (\y);
|
|
||||||
|
|
||||||
\foreach \y in {Yijp,Yipjp}
|
|
||||||
\draw[directed] (Wjp) -- (\y);
|
|
||||||
|
|
||||||
\foreach \y in {Yij,Yijp,Yipj,Yipjp}
|
|
||||||
\draw[directed] (alpha) -- (\y);
|
|
||||||
|
|
||||||
%--------------------------------------------------
|
%--------------------------------------------------
|
||||||
% Correlation structure
|
|
||||||
|
|
||||||
|
|
||||||
|
\node[observed] (Yij)
|
||||||
|
|
||||||
|
at ($(Zi)!0.5!(Wj)+(0,-0.9)$)
|
||||||
|
|
||||||
|
{$Y_{ij}$};
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
\draw[directed] (Zi) -- (Yij);
|
||||||
|
|
||||||
|
\draw[directed] (Wj) -- (Yij);
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
\draw[bidirected] (Zi) -- (Wj);
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
%--------------------------------------------------
|
%--------------------------------------------------
|
||||||
|
|
||||||
|
% Intercept
|
||||||
|
|
||||||
|
%--------------------------------------------------
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
\node[prior] (alpha)
|
||||||
|
|
||||||
|
at ($(Yij)+(0,-2.5)$)
|
||||||
|
|
||||||
|
{$\alpha$};
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
\draw[directed] (alpha) -- (Yij);
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\draw[bidirected] (alpha) -- (Zi);
|
\draw[bidirected] (alpha) -- (Zi);
|
||||||
\draw[bidirected] (alpha) -- (Zip);
|
|
||||||
|
|
||||||
\draw[bidirected] (alpha) -- (Wj);
|
\draw[bidirected] (alpha) -- (Wj);
|
||||||
\draw[bidirected] (alpha) -- (Wjp);
|
|
||||||
|
|
||||||
\end{tikzpicture}
|
|
||||||
\end{document}
|
|
||||||
```
|
|
||||||
|
|
||||||
```tikz
|
|
||||||
\usepackage{tikz}
|
|
||||||
|
|
||||||
\usepackage{amsmath}
|
%--------------------------------------------------
|
||||||
|
|
||||||
\usetikzlibrary{positioning,shapes.arrows, arrows.meta,shapes.geometric}
|
% Plates
|
||||||
|
|
||||||
\begin{document}
|
%--------------------------------------------------
|
||||||
|
|
||||||
\begin{tikzpicture}
|
|
||||||
|
|
||||||
\tikzset{
|
|
||||||
|
|
||||||
every path/.append style = {
|
\node[plate,fit=(Pi)(Zi),label={[align=center]left:$i=1,\ldots,n_1$}] {};
|
||||||
|
|
||||||
arrows = ->,
|
|
||||||
|
|
||||||
> = stealth,},
|
|
||||||
|
|
||||||
every node/.append style = {
|
\node[plate,fit=(Wj),label={[align=center]right:$j=1,\ldots,n_2$}] {};
|
||||||
|
|
||||||
shape = circle,
|
|
||||||
|
|
||||||
draw = black,
|
|
||||||
|
|
||||||
minimum size=3em
|
\node[plate,
|
||||||
|
|
||||||
},
|
fit=(Yij),
|
||||||
|
|
||||||
latent/.style = {
|
label=below right:{$(i,j)$}] {};
|
||||||
|
|
||||||
fill = lightgray
|
|
||||||
|
|
||||||
},
|
|
||||||
|
|
||||||
prior/.style = {
|
|
||||||
|
|
||||||
fill = red},
|
|
||||||
|
|
||||||
moral/.style = {
|
|
||||||
|
|
||||||
dashed,
|
|
||||||
|
|
||||||
> = {}, % remove arrow tip
|
|
||||||
|
|
||||||
arrows = -, % ensure no arrows
|
|
||||||
|
|
||||||
}}
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\node (y) {$Y$};
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\node[latent] (z) [above left = of y] {$Z$};
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\node[latent] (w) [above right = of y] {$W$};
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\node[latent] (P) [above = of z] {$P$};
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\node[prior] (sigma2) [above = of P] {$\sigma^2$};
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\node[prior] (rho) [above = of w] {$\rho_{1:R}$};
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\node[prior] (alpha) [below = of y] {$\boldsymbol{\alpha}$};
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\path (z) edge (y);
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\path (w) edge (y);
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\path (rho) edge (w);
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\path (alpha) edge (y);
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\path (P) edge (z);
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\path (sigma2) edge (P);
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
% moral
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\path[moral] (z) edge (alpha);
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\path[moral] (w) edge (alpha);
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
\path[moral] (z) edge (w);
|
|
||||||
|
|
||||||
\end{tikzpicture}
|
\end{tikzpicture}
|
||||||
|
|
||||||
|
|
@ -645,14 +600,14 @@ $$
|
||||||
\end{align*}
|
\end{align*}
|
||||||
$$
|
$$
|
||||||
|
|
||||||
En posant $R_{ir}=\sum_{j=1}^{n_{2}}W_{jr}Y_{ij}$ et $F_{ir}=\sum_{j=1}^{n_{2}}W_{jr}(1-Y_{ij})$ on définit les matrices $\mathbf{R}$ et $\mathbf{F}$ qui comptent les succès et échecs par ligne $i$ et groupe $r$.
|
En posant $R_{ir}^W=\sum_{j=1}^{n_{2}}W_{jr}Y_{ij}$ et $F_{ir}^W =\sum_{j=1}^{n_{2}}W_{jr}(1-Y_{ij})$ on définit les matrices $\mathbf{R}^W$ et $\mathbf{F}^W$ qui comptent les succès et échecs par ligne $i$ et groupe $r$.
|
||||||
|
|
||||||
Ce qui donne pour les $\tilde{\pi}_{i,k}$ de la posterior:
|
Ce qui donne pour les $\tilde{\pi}_{i,k}$ de la posterior:
|
||||||
|
|
||||||
$$
|
$$
|
||||||
\begin{align*}
|
\begin{align*}
|
||||||
\tilde{\pi}_{i,k} = p(Z_{i} = k\mid Y_{i,\bullet},\alpha,W,P_{i}) & \propto p(Y_{i,\bullet}\mid Z_{i} = k, \alpha, W, P_{i})p(Z_{i}=k\mid P_{i})\\
|
\tilde{\pi}_{i,k} = p(Z_{i} = k\mid Y_{i,\bullet},\alpha,W,P_{i}) & \propto p(Y_{i,\bullet}\mid Z_{i} = k, \alpha, W, P_{i})p(Z_{i}=k\mid P_{i})\\
|
||||||
& \propto \pi_{i,k} \prod_{r=1}^{R} \alpha_{k,r}^{R_{ir}}(1-\alpha_{k,r})^{F_{ir}}
|
& \propto \pi_{i,k} \prod_{r=1}^{R} \alpha_{k,r}^{R^W_{ir}}(1-\alpha_{k,r})^{F^W_{ir}}
|
||||||
\end{align*}
|
\end{align*}
|
||||||
$$
|
$$
|
||||||
|
|
||||||
|
|
@ -866,35 +821,33 @@ p(Y_{\bullet,j}\mid W_j=r,\alpha,Z)&= \prod_{i=1}^{n_1} \alpha_{Z_i,r}^{Y_{i,j}}
|
||||||
\end{align*}
|
\end{align*}
|
||||||
$$
|
$$
|
||||||
|
|
||||||
**A MODIFIER**
|
On peut poser $R_{jq}^{Z} = \sum_{i=1}^{n_{1}}\mathbb{1}_{Z_{i}=q} Y_{i,j}$ et $F_{jq}^{Z} = \sum_{i=1}^{n_{1}}\mathbb{1}_{Z_{i}=q} (1-Y_{i,j})$ et on définit les matrices $\mathbf{R}^Z$ et $\mathbf{F}^{Z}$.
|
||||||
|
|
||||||
On a :
|
|
||||||
|
|
||||||
$$
|
$$
|
||||||
\begin{align*}
|
\begin{align*}
|
||||||
p(Z_{i}\mid P_{i}) & = \ilr^{-1}(P_{i}) = (\pi_{i,1},\dots,\pi_{i,k},\dots,\pi_{i,K}) \\
|
p(W_{j}=r\mid Y_{\bullet,j},\alpha, Z, \rho) \propto \rho_{r}\prod_{q=1}^Q\alpha_{q,r}^{R^Z_{jq}}(1-\alpha_{q,r})^{F_{jq}^Z}
|
||||||
p(Y_{i,\bullet}\mid Z_{i}, \alpha, W) & = \prod_{j=1}^{n_{2}} \alpha_{Z_{i},W_{j}}^{Y_{ij}}(1- \alpha_{Z_{i},W_{j}})^{1-Y_{ij}} \\
|
\end{align*}
|
||||||
p(Z_{i} = k \mid P_{i}) & = \pi_{i,k} \\
|
|
||||||
p(Y_{i,\bullet}\mid Z_{i} = k, \alpha, W) & = \prod_{j=1}^{n_{2}} \prod_{r=1}^{R}\alpha_{k,r}^{\mathbb{1}_{W_{j} = r} Y_{ij}}(1- \alpha_{k,r})^{\mathbb{1}_{W_{j} = r}(1-Y_{ij})} \\
|
|
||||||
& = \prod_{r=1}^{R} \alpha_{k,r}^{\sum_{j=1}^{n_{2}}W_{jr}Y_{ij}} (1-\alpha_{k,r})^{\sum_{j=1}^{n_{2}}W_{jr}(1-Y_{ij})}
|
|
||||||
\end{align*}
|
|
||||||
$$
|
$$
|
||||||
|
|
||||||
En posant $R_{ir}=\sum_{j=1}^{n_{2}}W_{jr}Y_{ij}$ et $F_{ir}=\sum_{j=1}^{n_{2}}W_{jr}(1-Y_{ij})$ on définit les matrices $\mathbf{R}$ et $\mathbf{F}$ qui comptent les succès et échecs par ligne $i$ et groupe $r$.
|
À la fin
|
||||||
|
|
||||||
Ce qui donne pour les $\tilde{\pi}_{i,k}$ de la posterior:
|
$$
|
||||||
|
W_{j} \sim \Cat_{R}(\tilde{\rho}_{1:R}^j)
|
||||||
|
$$
|
||||||
|
|
||||||
|
### Implémentation
|
||||||
|
|
||||||
$$
|
$$
|
||||||
\begin{align*}
|
\begin{align*}
|
||||||
\tilde{\pi}_{i,k} = p(Z_{i} = k\mid Y_{i,\bullet},\alpha,W,P_{i}) & \propto p(Y_{i,\bullet}\mid Z_{i} = k, \alpha, W, P_{i})p(Z_{i}=k\mid P_{i})\\
|
\log \tilde{p_{jr}} & = \log \rho_{r} + \sum_{q=1}^{Q} [R_{jq}^{Z} \log\alpha_{q,r} + F_{jq}^W\log{(1-\alpha_{q,r})}] \\
|
||||||
& \propto \pi_{i,k} \prod_{r=1}^{R} \alpha_{k,r}^{R_{ir}}(1-\alpha_{k,r})^{F_{ir}}
|
\tilde{\rho}_{jr} &= \frac{\exp(\log \tilde{p}_{jr} - m_{j})}{\sum_{l=1}^{R}\exp(\log \tilde{p_{jr}}-m_{j})},\quad m_{j} = \max_{l} \log p_{jr}
|
||||||
\end{align*}
|
\end{align*}
|
||||||
$$
|
$$
|
||||||
|
|
||||||
Et ainsi à la fin :
|
Ainsi :
|
||||||
|
|
||||||
$$
|
$$
|
||||||
Z_{i}\mid P_{i}, Y, W, \alpha \sim \Cat_{K}(\tilde{\pi}_{i,1},\dots, \tilde{\pi}_{i,K})
|
\log \tilde{R} = \log(\rho) + \mathbf{R}^Z\log\alpha^{\top} + \mathbf{F}^Z \log(1-\alpha)^{\top}
|
||||||
$$
|
$$
|
||||||
|
|
||||||
## Loi de $\alpha \mid Y,Z,W$
|
## Loi de $\alpha \mid Y,Z,W$
|
||||||
|
|
|
||||||
|
|
@ -18,4 +18,6 @@ Exécution arrêtée
|
||||||
```
|
```
|
||||||
|
|
||||||
- [ ] Essayer de reproduire l'erreur de `vapply` pour la réparer 🆔 ckx0ew
|
- [ ] Essayer de reproduire l'erreur de `vapply` pour la réparer 🆔 ckx0ew
|
||||||
|
+ Elle semble lier au fait d'ajouter des modèles `NULL`
|
||||||
|
+ [ ] Ajouter la même suppression pour discarded que pour compared
|
||||||
- [ ] Implémenter un test unitaire pour prévenir la régression ⛔ ckx0ew
|
- [ ] Implémenter un test unitaire pour prévenir la régression ⛔ ckx0ew
|
||||||
|
|
|
||||||
Loading…
Add table
Reference in a new issue