diff --git a/lab2/solutions/Part2_Debiasing_Solution.ipynb b/lab2/solutions/Part2_Debiasing_Solution.ipynb index cd2b5888..aef6eff1 100644 --- a/lab2/solutions/Part2_Debiasing_Solution.ipynb +++ b/lab2/solutions/Part2_Debiasing_Solution.ipynb @@ -21,19 +21,19 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": null, "metadata": { "id": "rNbf1pRlSDby" }, "outputs": [], "source": [ - "# Copyright 2023 MIT 6.S191 Introduction to Deep Learning. All Rights Reserved.\n", + "# Copyright 2023 MIT Introduction to Deep Learning. All Rights Reserved.\n", "# \n", "# Licensed under the MIT License. You may not use this file except in compliance\n", - "# with the License. Use and/or modification of this code outside of 6.S191 must\n", - "# reference:\n", + "# with the License. Use and/or modification of this code outside of MIT Introduction\n", + "# to Deep Learning must reference:\n", "#\n", - "# © MIT 6.S191: Introduction to Deep Learning\n", + "# © MIT Introduction to Deep Learning\n", "# http://introtodeeplearning.com\n", "#" ] @@ -46,13 +46,16 @@ "source": [ "# Laboratory 2: Computer Vision\n", "\n", - "# Part 2: Debiasing Facial Detection Systems\n", + "# Part 2: Facial Detection Systems\n", + "\n", + "In the second portion of Lab 2, we'll explore a prominent aspect of applied deep learning for computer vision: facial detection. \n", "\n", - "In the second portion of the lab, we'll explore a prominent aspect of applied deep learning: facial detection. \n", + "Consider the task of facial detection: given an image, is it an image of a face? This seemingly simple -- but extremely important and pervasive -- task is subject to significant amounts of algorithmic bias among select demographics, as [seminal studies](https://proceedings.mlr.press/v81/buolamwini18a/buolamwini18a.pdf) have shown.\n", "\n", - "Deploying fair, unbiased AI systems is critical to their long-term acceptance. Consider the task of facial detection: given an image, is it an image of a face? This seemingly simple, but extremely important, task is subject to significant amounts of algorithmic bias among select demographics. \n", + "Deploying fair, unbiased AI systems is critical to their long-term acceptance. In this lab, we will build computer vision models for facial detection. We will extend beyond that to build a model to **uncover and diagnose** the biases and issues that exist with standard facial detection models. To do this, we will build a semi-supervised variational autoencoder (SS-VAE) that learns the *latent distribution* of features underlying face image datasets in order to [uncover hidden biases](http://introtodeeplearning.com/AAAI_MitigatingAlgorithmicBias.pdf).\n", "\n", - "In this lab, we'll start by building a variational autoencoder that learns the latent distribution of features in a popular facial recognition dataset, with the goal of debiasing using these latents usine [one recently published approach](http://introtodeeplearning.com/AAAI_MitigatingAlgorithmicBias.pdf) in the next lab. We'll build a facial detection model that learns the *latent variables* underlying face image datasets and uses this to adaptively re-sample the training data, thus mitigating any biases that may be present in order to train a *debiased* model." + "Our work here will set the foundation for Lab 3, where we'll build automated tools to mitigate the underlying issues of bias and uncertainty in facial detection.\n", + "\n" ] }, { @@ -66,10 +69,10 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 1, "metadata": { "id": "E46sWVKK6LP9", - "outputId": "38a5e4f9-3541-41a4-f861-a8965a00536e", + "outputId": "4443314f-d922-49d9-d0f2-5fc6c0e6a41a", "colab": { "base_uri": "https://localhost:8080/" } @@ -81,21 +84,31 @@ "text": [ "Colab only includes TensorFlow 2.x; %tensorflow_version has no effect.\n", "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n", - "Requirement already satisfied: mitdeeplearning in /usr/local/lib/python3.8/dist-packages (0.2.0)\n", - "Requirement already satisfied: tqdm in /usr/local/lib/python3.8/dist-packages (from mitdeeplearning) (4.64.1)\n", + "Collecting mitdeeplearning\n", + " Downloading mitdeeplearning-0.2.0.tar.gz (2.1 MB)\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.1/2.1 MB\u001b[0m \u001b[31m44.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", + "\u001b[?25h Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n", "Requirement already satisfied: numpy in /usr/local/lib/python3.8/dist-packages (from mitdeeplearning) (1.21.6)\n", "Requirement already satisfied: regex in /usr/local/lib/python3.8/dist-packages (from mitdeeplearning) (2022.6.2)\n", + "Requirement already satisfied: tqdm in /usr/local/lib/python3.8/dist-packages (from mitdeeplearning) (4.64.1)\n", "Requirement already satisfied: gym in /usr/local/lib/python3.8/dist-packages (from mitdeeplearning) (0.25.2)\n", + "Requirement already satisfied: cloudpickle>=1.2.0 in /usr/local/lib/python3.8/dist-packages (from gym->mitdeeplearning) (1.5.0)\n", "Requirement already satisfied: gym-notices>=0.0.4 in /usr/local/lib/python3.8/dist-packages (from gym->mitdeeplearning) (0.0.8)\n", "Requirement already satisfied: importlib-metadata>=4.8.0 in /usr/local/lib/python3.8/dist-packages (from gym->mitdeeplearning) (5.2.0)\n", - "Requirement already satisfied: cloudpickle>=1.2.0 in /usr/local/lib/python3.8/dist-packages (from gym->mitdeeplearning) (1.5.0)\n", - "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.8/dist-packages (from importlib-metadata>=4.8.0->gym->mitdeeplearning) (3.11.0)\n" + "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.8/dist-packages (from importlib-metadata>=4.8.0->gym->mitdeeplearning) (3.11.0)\n", + "Building wheels for collected packages: mitdeeplearning\n", + " Building wheel for mitdeeplearning (setup.py) ... \u001b[?25l\u001b[?25hdone\n", + " Created wheel for mitdeeplearning: filename=mitdeeplearning-0.2.0-py3-none-any.whl size=2115441 sha256=68cd8e46eb0b3de78bd7f6db6c547003ffcbd6a5aaecb212f283910d0d1ac74e\n", + " Stored in directory: /root/.cache/pip/wheels/2e/45/44/c5b304f31f37e8d2315f9e969fd8cdb0014a5c28608d0bf410\n", + "Successfully built mitdeeplearning\n", + "Installing collected packages: mitdeeplearning\n", + "Successfully installed mitdeeplearning-0.2.0\n" ] } ], "source": [ "# Import Tensorflow 2.0\n", - "%tensorflow_version 2.9\n", + "%tensorflow_version 2.x\n", "import tensorflow as tf\n", "\n", "import IPython\n", @@ -105,7 +118,7 @@ "from tqdm import tqdm\n", "\n", "\n", - "# Download and import the MIT 6.S191 package\n", + "# Download and import the MIT Introduction to Deep Learning package\n", "!pip install mitdeeplearning\n", "import mitdeeplearning as mdl" ] @@ -129,10 +142,10 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 2, "metadata": { "id": "RWXaaIWy6jVw", - "outputId": "8e2f99d8-4e0d-4cea-c915-84f400fce78b", + "outputId": "d49ebf8d-5862-4244-d71c-e9082030aed8", "colab": { "base_uri": "https://localhost:8080/" } @@ -143,7 +156,7 @@ "name": "stdout", "text": [ "Downloading data from https://www.dropbox.com/s/hlz8atheyozp1yx/train_face.h5?dl=1\n", - "1263889489/1263889489 [==============================] - 50s 0us/step\n", + "1263889489/1263889489 [==============================] - 12s 0us/step\n", "Opening /root/.keras/datasets/train_face.h5\n", "Loading data into memory...\n" ] @@ -167,7 +180,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 3, "metadata": { "id": "DjPSjZZ_bGqe" }, @@ -188,13 +201,13 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 11, "metadata": { "id": "Jg17jzwtbxDA", - "outputId": "6f4331d3-5b11-456b-bab0-a3ade424edd8", + "outputId": "82a2c9ad-09dd-4f1b-b1b2-f5b25b8ff676", "colab": { "base_uri": "https://localhost:8080/", - "height": 191 + "height": 267 } }, "outputs": [ @@ -202,9 +215,9 @@ "output_type": "display_data", "data": { "text/plain": [ - "
" + "
" ], - "image/png": "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\n" + "image/png": "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\n" }, "metadata": { "needs_background": "light" @@ -220,9 +233,9 @@ "not_face_images = images[np.where(labels==0)[0]]\n", "\n", "idx_face = 29 #@param {type:\"slider\", min:0, max:50, step:1}\n", - "idx_not_face = 9 #@param {type:\"slider\", min:0, max:50, step:1}\n", + "idx_not_face = 33 #@param {type:\"slider\", min:0, max:50, step:1}\n", "\n", - "plt.figure(figsize=(5,5))\n", + "plt.figure(figsize=(8,4))\n", "plt.subplot(1, 2, 1)\n", "plt.imshow(face_images[idx_face])\n", "plt.title(\"Face\"); plt.grid(False)\n", @@ -262,7 +275,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": null, "metadata": { "id": "82EVTAAW7B_X" }, @@ -312,7 +325,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": null, "metadata": { "id": "eJlDGh1o31G1", "outputId": "ef3b8a20-ec73-4978-878f-67f84b3bd8cd", @@ -415,7 +428,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": null, "metadata": { "id": "35-PDgjdWk6_", "outputId": "3babb0e6-09a1-4970-b4da-b4c09f8660e1", @@ -458,7 +471,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": null, "metadata": { "id": "vfDD8ztGWk6x", "outputId": "8e27976c-8f6d-43c6-df11-b45b98d4da5f", @@ -539,7 +552,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": null, "metadata": { "id": "GI4O0Y1GAot9", "outputId": "e4416a91-50a5-4b17-ded0-39c418dff13c", @@ -659,7 +672,7 @@ }, { "cell_type": "code", - "execution_count": 28, + "execution_count": null, "metadata": { "id": "S00ASo1ImSuh" }, @@ -725,7 +738,7 @@ }, { "cell_type": "code", - "execution_count": 29, + "execution_count": null, "metadata": { "id": "cT6PGdNajl3K" }, @@ -800,16 +813,16 @@ "\n", "In contrast, for images of **non-faces**, our loss function is solely the classification loss. \n", "\n", - "We can write a single expression for the loss by defining an indicator variable $\\mathcal{I}_f$which reflects which training data are images of faces ($\\mathcal{I}_f(y) = 1$ ) and which are images of non-faces ($\\mathcal{I}_f(y) = 0$). Using this, we obtain:\n", + "We can write a single expression for the loss by defining an indicator variable ${I}_f$which reflects which training data are images of faces (${I}_f(y) = 1$ ) and which are images of non-faces (${I}_f(y) = 0$). Using this, we obtain:\n", "\n", - "$$L_{total} = L_y(y,\\hat{y}) + \\mathcal{I}_f(y)\\Big[L_{VAE}\\Big]$$\n", + "$$L_{total} = L_y(y,\\hat{y}) + {I}_f(y)\\Big[L_{VAE}\\Big]$$\n", "\n", "Let's write a function to define the DB-VAE loss function:\n" ] }, { "cell_type": "code", - "execution_count": 30, + "execution_count": null, "metadata": { "id": "VjieDs8Ovcqs" }, @@ -870,7 +883,7 @@ }, { "cell_type": "code", - "execution_count": 31, + "execution_count": null, "metadata": { "id": "JfWPHGrmyE7R" }, @@ -916,7 +929,7 @@ }, { "cell_type": "code", - "execution_count": 32, + "execution_count": null, "metadata": { "id": "dSFDcFBL13c3" }, @@ -951,16 +964,9 @@ "\n", " return y_logit, z_mean, z_logsigma\n", "\n", - " # VAE reparameterization: given a mean and logsigma, sample latent variables\n", - " def reparameterize(self, z_mean, z_logsigma):\n", - " # TODO: call the sampling function defined above\n", - " z = sampling(z_mean, z_logsigma)\n", - " # z = # TODO\n", - " return z\n", - "\n", " # Decode the latent space and output reconstruction\n", " def decode(self, z):\n", - " # TODO: use the decoder to output the reconstruction\n", + " # TODO: use the decoder (self.decoder) to output the reconstruction\n", " reconstruction = self.decoder(z)\n", " # reconstruction = # TODO\n", " return reconstruction\n", @@ -970,8 +976,9 @@ " # Encode input to a prediction and latent space\n", " y_logit, z_mean, z_logsigma = self.encode(x)\n", "\n", - " # TODO: reparameterization\n", - " z = self.reparameterize(z_mean, z_logsigma)\n", + " # TODO: call the sampling function that you created above using \n", + " # z_mean and z_logsigma\n", + " z = sampling(z_mean, z_logsigma)\n", " # z = # TODO\n", "\n", " # TODO: reconstruction\n", @@ -1011,7 +1018,7 @@ }, { "cell_type": "code", - "execution_count": 156, + "execution_count": null, "metadata": { "id": "xwQs-Gu5bKEK", "outputId": "e05cae22-ba7a-4598-ec3e-f19f2a6f6952", @@ -1252,7 +1259,7 @@ }, { "cell_type": "code", - "execution_count": 384, + "execution_count": null, "metadata": { "id": "bgK77aB9oDtX", "outputId": "209dc35f-ef86-4740-b932-1cd2cf685adf", @@ -1302,7 +1309,7 @@ }, { "cell_type": "code", - "execution_count": 347, + "execution_count": null, "metadata": { "id": "BRAwm0Jlrktc", "outputId": "d0645ef2-7030-4f70-e0f3-87e4154eb344", @@ -1370,18 +1377,6 @@ "As we've seen above, loss is a powerful way to visualize which samples in our dataset have high *uncertainty*, or which ones the model has had trouble learning. However, this isn't necessarily the same as bias! How can determine the *probability* of a sample occurring in our dataset, and debias based off of that? In this section, we'll develop a way to score samples based on their bias and adapt this score during training." ] }, - { - "cell_type": "code", - "source": [ - "# (x, y) = loader.get_batch(5000, only_faces=True)\n", - "# y_logit, z_mean, z_logsigma, x_recon = dbvae(x)" - ], - "metadata": { - "id": "zZX4jlCRDkuq" - }, - "execution_count": 348, - "outputs": [] - }, { "cell_type": "code", "source": [ @@ -1417,7 +1412,7 @@ "height": 258 } }, - "execution_count": 383, + "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -1436,9 +1431,6 @@ { "cell_type": "code", "source": [ - "# (x, y) = loader.get_batch(5000, only_faces=True)\n", - "# y_logit, z_mean, z_logsigma, x_recon = dbvae(x)\n", - "\n", "avg_logit_per_bin = []\n", "for idx_latent in range(latent_dim): \n", " latent_samples = z_mean[:, idx_latent]\n", @@ -1469,7 +1461,7 @@ "height": 279 } }, - "execution_count": 371, + "execution_count": null, "outputs": [ { "output_type": "display_data",