{ "cells": [ { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "# Clustering with Gaussian Mixture Models" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Preliminaries\n", "\n", "- Goal \n", " - Introduction to the Expectation-Maximization (EM) Algorithm with application to Gaussian Mixture Models (GMM)\n", "- Materials \n", " - Mandatory\n", " - These lecture notes\n", " - Optional\n", " - Bishop pp. 430-439 for Gaussian Mixture Models" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Limitations of Simple IID Gaussian Models\n", "\n", "Sofar, model inference was solved analytically, but we used strong assumptions:\n", "\n", "- IID sampling, $p(D) = \\prod_n p(x_n)$" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- Simple Gaussian (or multinomial) PDFs, $p(x_n) = \\mathcal{N}(x_n|\\mu,\\Sigma)$" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- Some limitations of Simple Gaussian Models with IID Sampling:\n", " - What if the PDF is **multi-modal** (or is just not Gaussian in any other way)?\n", " - Covariance matrix $\\Sigma$ has $D(D+1)/2$ parameters.\n", " - This quickly becomes **a very large number** for increasing dimension $D$.\n", " - Temporal signals are often **not IID**." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Towards More Flexible Models\n", "\n", "- What if the PDF is multi-modal (or is just not Gaussian in any other way)?\n", " - **Discrete latent** variable models (a.k.a. **mixture** models). We'll cover this case in this lesson.\n", " " ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ " \n", "- Covariance matrix $\\Sigma$ has $D(D+1)/2$ parameters. This quickly becomes very large for increasing dimension $D$.\n", " - **Continuous latent** variable models (a.k.a. **dimensionality reduction** models). Covered in [lesson 11](http://nbviewer.ipython.org/github/bertdv/AIP-5SSB0/blob/master/lessons/notebooks/11_Continuous-Latent-Variable-Models-PCA-and-FA.ipynb)." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- Temporal signals are often not IID.\n", " - Introduce **Markov dependencies** and **latent state** variable models. This will be covered in [lesson 13](http://nbviewer.jupyter.org/github/bertdv/AIP-5SSB0/blob/master/lessons/notebooks/13_Dynamic-Latent-Variable-Models.ipynb).\n", " " ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Illustrative Example\n", "\n", "- You're now asked to build a density model for a data set ([Old Faithful](https://en.wikipedia.org/wiki/Old_Faithful), Bishop pg. 681) that clearly is not distributed as a single Gaussian:\n", "\n", "" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Unobserved Classes\n", "\n", "Consider again a set of observed data $D=\\{x_1,\\dotsc,x_N\\}$\n", "\n", "- This time we suspect that there are unobserved class labels that would help explain (or predict) the data, e.g.,\n", " - the observed data are the color of living things; the unobserved classes are animals and plants.\n", " - observed are wheel sizes; unobserved categories are trucks and personal cars.\n", " - observed is an audio signal; unobserved classes include speech, music, traffic noise, etc.\n", " " ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ " \n", "- Classification problems with unobserved classes are called **Clustering** problems. The learning algorithm needs to **discover the classes from the observed data**." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### The Gaussian Mixture Model (GMM)\n", " \n", "- If the categories were observed as well, these data could be nicely modeled by the previously discussed [generative classification framework](http://nbviewer.jupyter.org/github/bertdv/AIP-5SSB0/blob/master/lessons/07_generative_classification/Generative-Classification.ipynb)." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- Let us use **equivalent model assumptions to linear generative classification**: \n", "$$\\begin{align*} p(x_n,\\mathcal{C}_k) &= \\overbrace{p(\\mathcal{C}_k)}^{\\text{class prior}} \\, \\overbrace{p(x_n|\\mathcal{C}_k)}^{\\text{likelihood}} \n", "= \\pi_k \\,\\mathcal{N}\\left(x_n|\\mu_k,\\Sigma_k \\right) \\end{align*}$$\n", "where, as previously, we use notational shorthand $\\mathcal{C}_k \\triangleq (z_{nk}=1)$ and a _hidden_ $1$-of-$K$ selector variable $z_{nk}$ with each observation $x_n$ as \n", "$$\n", "z_{nk} = \\begin{cases} 1 & \\text{if } \\, z_n \\in \\mathcal{C}_k\\\\\n", "0 & \\text{otherwise} \\end{cases}\n", "$$" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- Since the class labels are not observed, the probability for observations in a GMM is obtained through _marginalization_ over the classes (this is different in classification problems, where the classes are observed):\n", "$$\\begin{align*}\n", "p(x_n) = \\boxed{\\sum_k \\pi_k \\mathcal{N}\\left(x_n|\\mu_k,\\Sigma_k \\right)}\n", "\\end{align*}$$\n", " - The class priors $p(\\mathcal{C}_k) =\\pi_k$ are often called _mixture coefficients_. " ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Gaussian Mixture Models\n", "\n", "- GMMs are very popular models. They have decent computational properties and are **universal approximators of densities** (as long as there are enough Gaussians of course)\n", "\n", "\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Inference: Log-Likelihood for GMM\n", "\n", "- The log-likelihood for observed data $D=\\{x_1,\\dotsc,x_N\\}$,\n", "$$\\begin{align*}\n", "\\log p(D|\\theta) &\\stackrel{\\text{IID}}{=} \\sum_n \\log p(x_n|\\theta)\n", " = \\sum_n \\log \\left( \\sum_{k=1}^K p(\\mathcal{C}_k) \\, p(x_n|\\mathcal{C}_k) \\right) \\\\\n", " &= \\sum_n \\log \\left( \\sum_k \\pi_k\\mathcal{N}(x_n|\\mu_k,\\Sigma_k) \\right)\n", "\\end{align*}$$\n", "... and now the log-of-sum cannot be further simplified." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- Compare this to the log-likelihood for $D=\\{(x_1,y_1),\\dotsc,(x_N,y_N)\\}$ in [(generative) classification](http://nbviewer.ipython.org/github/bertdv/AIP-5SSB0/blob/master/lessons/notebooks/07_Generative-Classification.ipynb): \n", "$$\\begin{align*}\n", "\\log p(D|\\theta) &= \\sum_n \\log \\prod_k p(x_n,\\mathcal{C}_{k}|\\theta)^{y_{nk}} \n", " = \\sum_{n,k} y_{nk} \\log p(x_n,\\mathcal{C}_{k}|\\theta) \\\\\n", " &= \\sum_k m_k \\log \\pi_k + \\sum_{n,k} y_{nk} \\log \\mathcal{N}(x_n|\\mu_k,\\Sigma)\n", "\\end{align*}$$\n", "which led to easy Gaussian and multinomial parameter estimation." ] }, { "cell_type": "markdown", "metadata": { "collapsed": true, "slideshow": { "slide_type": "fragment" } }, "source": [ "- Fortunately GMMs can be trained by maximum likelihood using an efficient algorithm: Expectation-Maximization." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Introducing a Soft Class Indicator\n", "\n", "Let's introduce a $1$-of-$K$ selector variable $z_{nk}$ to represent the _unobserved_ classes $\\mathcal{C}_k$ by \n", "$$\n", "z_{nk} = \\begin{cases} 1 & \\text{if $z_n$ in class $\\mathcal{C}_k$}\\\\\n", " 0 & \\text{otherwise} \\end{cases}\n", "$$" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- We don't _observe_ the class labels $z_{nk}$, but we can compute the posterior probability for the class labels, given the observation $x_n$:\n", "$$\\begin{align*}\n", "p(\\mathcal{C}_k | x_n ) &= \\frac{p(x_n|\\mathcal{C}_k)p(\\mathcal{C}_k)}{\\sum_{k^\\prime} p(x_n|\\mathcal{C}_{k^\\prime})p(\\mathcal{C}_{k^\\prime})} \n", " = \\frac{\\pi_k \\mathcal{N}(x_n | \\mu_k,\\Sigma_k)}{\\sum_{k^\\prime} \\pi_{k^\\prime} \\mathcal{N}(x_n | \\mu_{k^\\prime},\\Sigma_{k^\\prime})}\n", "\\end{align*}$$" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- The $\\gamma_{nk} \\triangleq p(\\mathcal{C}_k | x_n ) = p(z_{nk}=1 | x_n )$ are also called **responsibilities**. Note that $0 \\leq \\gamma_{nk} \\leq 1$ by definition and that they can be evaluated (i.e., we can compute it)." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- The responsibilities $\\gamma_{nk}$ are soft class indicators and they play the same role in clustering as class selection variables ($y_{nk}$) do in classification problems. " ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### ML estimation for Clustering: The Expectation-Maximization (EM) Algorithm Idea \n", "\n", "- IDEA: Let's apply the (generative) classification formulas and substitute the reponsibilities $\\gamma_{nk}$ wherever the formulas use the binary class indicators $y_{nk}$." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- Try parameter updates (same as for [generative Gaussian classification](http://nbviewer.ipython.org/github/bertdv/AIP-5SSB0/blob/master/lessons/notebooks/07_Generative-Classification.ipynb)):\n", "$$\\begin{align*}\n", "\\hat \\pi_k &= \\frac{m_k}{N}\\,,\\quad \\text{where }m_k = \\sum_n \\gamma_{nk} \\\\\n", "\\hat \\mu_k &= \\frac{1}{m_k} \\sum_n \\gamma_{nk} x_n \\\\\n", "\\hat \\Sigma_k &= \\frac{1}{m_k} \\sum_{n} \\gamma_{nk} (x_n-\\hat \\mu_k)(x_n-\\hat \\mu_k)^T\n", "\\end{align*}$$" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- But wait ..., the responsibilities $\\gamma_{nk}=\\frac{\\pi_k \\mathcal{N}(x_n|\\mu_k,\\Sigma_k)}{\\sum_j \\pi_j \\mathcal{N}(x_n|\\mu_j,\\Sigma_j)}$ are a function of the model parameters $\\{\\pi,\\mu,\\Sigma\\}$ and the parameter updates depend on the responsibilities ..." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- **Solution**: Iterate updating between $\\gamma_{nk}$ and the parameters $\\{\\pi,\\mu,\\Sigma\\}$. This procedure is called the **Expectation-Maximization (EM)** algorithm." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "#### CODE EXAMPLE\n", "\n", "We'll perform clustering on the data set from the illustrative example by fitting a GMM consisting of two Gaussians using the EM algorithm. " ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "Figure(PyObject
)" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "using DataFrames, CSV, LinearAlgebra\n", "include(\"scripts/gmm_plot.jl\") # Holds plotting function \n", "old_faithful = CSV.read(\"datasets/old_faithful.csv\")\n", "X = convert(Matrix{Float64}, [old_faithful[1] old_faithful[2]]')\n", "N = size(X, 2)\n", "\n", "# Initialize the GMM. We assume 2 clusters.\n", "clusters = [MvNormal([4.;60.], [.5 0;0 10^2]); \n", " MvNormal([2.;80.], [.5 0;0 10^2])]\n", "π_hat = [0.5; 0.5] # Mixing weights\n", "γ = fill!(Matrix{Float64}(undef,2,N), NaN) # Responsibilities (row per cluster)\n", "\n", "# Define functions for updating the parameters and responsibilities\n", "function updateResponsibilities!(X, clusters, π_hat, γ)\n", " # Expectation step: update γ\n", " norm = [pdf(clusters[1], X) pdf(clusters[2], X)] * π_hat\n", " γ[1,:] = (π_hat[1] * pdf(clusters[1],X) ./ norm)'\n", " γ[2,:] = 1 .- γ[1,:]\n", "end\n", "function updateParameters!(X, clusters, π_hat, γ)\n", " # Maximization step: update π_hat and clusters using ML estimation\n", " m = sum(γ, dims=2)\n", " π_hat = m / N\n", " μ_hat = (X * γ') ./ m'\n", " for k=1:2\n", " Z = (X .- μ_hat[:,k])\n", " Σ_k = Symmetric(((Z .* (γ[k,:])') * Z') / m[k])\n", " clusters[k] = MvNormal(μ_hat[:,k], convert(Matrix, Σ_k))\n", " end\n", "end\n", "\n", "# Execute the algorithm: iteratively update parameters and responsibilities\n", "subplot(2,3,1); plotGMM(X, clusters, γ); title(\"Initial situation\")\n", "updateResponsibilities!(X, clusters, π_hat, γ)\n", "subplot(2,3,2); plotGMM(X, clusters, γ); title(\"After first E-step\")\n", "updateParameters!(X, clusters, π_hat, γ)\n", "subplot(2,3,3); plotGMM(X, clusters, γ); title(\"After first M-step\")\n", "iter_counter = 1\n", "for i=1:3\n", " for j=1:i+1\n", " updateResponsibilities!(X, clusters, π_hat, γ)\n", " updateParameters!(X, clusters, π_hat, γ)\n", " iter_counter += 1\n", " end\n", " subplot(2,3,3+i); \n", " plotGMM(X, clusters, γ); \n", " title(\"After $(iter_counter) iterations\")\n", "end\n", "PyPlot.tight_layout()" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "Note that you can step through the interactive demo yourself by running [this script](https://github.com/bertdv/AIP-5SSB0/blob/master/lessons/notebooks/scripts/interactive_em_demo.jl) in julia. You can run a script in julia by \n", "`julia> include(\"path/to/script-name.jl\")`" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Clustering vs. (Generative) Classification\n", "\n", "\n", " \n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", " \n", "\n", "\n", "\n", "
Classification Clustering
1Class label $y_n$ is observed Class label $z_n$ is latent
2log-likelihood conditions on observed class
$$\\propto \\sum_{nk} y_{nk} \\log \\mathcal{N}(x_n|\\mu_k,\\Sigma_k)$$
log-likelihood marginalizes over latent classes
$$\\propto \\sum_{n}\\log \\sum_k \\pi_k \\mathcal{N}(x_n|\\mu_k,\\Sigma_k)$$
3'Hard' class selector
$$\n", "y_{nk} = \\begin{cases} 1 & \\text{if $y_n$ in class $\\mathcal{C}_k$}\\\\\n", " 0 & \\text{otherwise} \\end{cases}\n", "$$
'Soft' class responsibility
$$\\gamma_{nk} = p(\\mathcal{C}_k|x_n)$$
4Estimation:
\n", "$$\\begin{align*}\n", "\\hat{\\pi}_k &= \\frac{1}{N}\\sum_n y_{nk} \\\\\n", "\\hat{\\mu}_k &= \\frac{\\sum_n y_{nk} x_n}{\\sum_n y_{nk}} \\\\\n", "\\hat{\\Sigma}_k &= \\frac{\\sum_n y_{nk} (x_n-\\hat\\mu_k)(x_n-\\hat\\mu_k)^T}{\\sum_n y_{nk}}\n", "\\end{align*}$$\n", "
Estimation (1 update of M-step!)
\n", "$$\\begin{align*}\n", "\\hat{\\pi}_k &= \\frac{1}{N}\\sum_n \\gamma_{nk} \\\\\n", "\\hat{\\mu}_k &= \\frac{\\sum_n \\gamma_{nk} x_n}{\\sum_n \\gamma_{nk}} \\\\\n", "\\hat{\\Sigma}_k &= \\frac{\\sum_n \\gamma_{nk} (x_n-\\hat\\mu_k)(x_n-\\hat\\mu_k)^T}{\\sum_n \\gamma_{nk}}\n", "\\end{align*}$$\n", "
" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "slideshow": { "slide_type": "skip" } }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "open(\"../../styles/aipstyle.html\") do f\n", " display(\"text/html\", read(f,String))\n", "end" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "anaconda-cloud": {}, "celltoolbar": "Slideshow", "kernelspec": { "display_name": "Julia 1.1.0", "language": "julia", "name": "julia-1.1" }, "language_info": { "file_extension": ".jl", "mimetype": "application/julia", "name": "julia", "version": "1.1.0" } }, "nbformat": 4, "nbformat_minor": 1 }