{ "cells": [ { "cell_type": "markdown", "id": "61c4bf28", "metadata": {}, "source": [ "# Tutorial 3: Solutions to Exercises 3.1 & 3.2\n", "\n", "![Status](https://img.shields.io/static/v1.svg?label=Status&message=Open&color=blue)\n", "\n", "\n", "**Open notebook on:** \n", "[![View filled on Github](https://img.shields.io/static/v1.svg?logo=github&label=Repo&message=View%20On%20Github&color=lightgrey)](https://github.com/clandolt/mlcysec_notebooks/blob/main/source/tutorial_notebooks/tutorial3_solution/tutorial3_solution.ipynb)\n", "[![Open filled In Collab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/clandolt/mlcysec_notebooks/blob/main/source/tutorial_notebooks/tutorial3_solution/tutorial3_solution.ipynb) \n", "**Author:** Christoph R. Landolt\n", "\n", "This notebook provides the solutions for the exercises corresponding to Tutorials 3.1 and 3.2." ] }, { "cell_type": "markdown", "id": "c0605e51", "metadata": {}, "source": [ "## Solutions for Tutorial 3.1: Getting started with GANs\n", "\n", "### Exercise 1: Experimenting with Architectural Changes\n", "\n", "#### 1. Implementation: High-Capacity Generator\n", "In this exercise, we investigate how increasing the capacity (the number of layers and neurons) of the Generator affects the training outcome when the Discriminator remains simple.\n", "\n", "We modified the `Generator` class to include an additional hidden layer and increased the width of the layers from 50 to 128 neurons.\n", "\n", "**Original Generator:**\n", "```python\n", "class Generator(nn.Module):\n", " def __init__(self):\n", " super(Generator, self).__init__()\n", " # Input dimension is 10 (noise), output dimension is 2 (for 2D data)\n", " self.fc = nn.Sequential(\n", " nn.Linear(10, 50),\n", " nn.ReLU(),\n", " nn.Linear(50, 2)\n", " )\n", "```\n", "**High-Capacity Generator:**\n", "```python\n", "class Generator(nn.Module):\n", " def __init__(self):\n", " super(Generator, self).__init__()\n", " # Input dimension is 10 (noise), output dimension is 2 (for 2D data)\n", " self.fc = nn.Sequential(\n", " nn.Linear(10, 50),\n", " nn.ReLU(),\n", " nn.Linear(50, 2)\n", " )\n", "```\n" ] }, { "cell_type": "code", "execution_count": 1, "id": "de0c88d5", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training with High-Capacity Generator on cpu...\n", "Epoch [0/10000] | Loss D: 2.3853 | Loss G: 0.6585\n", "Epoch [1000/10000] | Loss D: 1.4730 | Loss G: 0.6600\n", "Epoch [2000/10000] | Loss D: 1.2944 | Loss G: 0.7798\n", "Epoch [3000/10000] | Loss D: 1.7352 | Loss G: 0.5906\n", "Epoch [4000/10000] | Loss D: 1.2874 | Loss G: 0.7426\n", "Epoch [5000/10000] | Loss D: 1.4147 | Loss G: 0.6181\n", "Epoch [6000/10000] | Loss D: 1.4654 | Loss G: 0.6584\n", "Epoch [7000/10000] | Loss D: 1.3330 | Loss G: 0.9366\n", "Epoch [8000/10000] | Loss D: 1.4036 | Loss G: 0.7082\n", "Epoch [9000/10000] | Loss D: 1.3616 | Loss G: 0.6705\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import torch\n", "import torch.nn as nn\n", "import torch.optim as optim\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "\n", "# Check for GPU\n", "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", "\n", "# --- 1. Define the Models ---\n", "\n", "class HighCapacityGenerator(nn.Module):\n", " def __init__(self, input_dim=10, output_dim=2):\n", " super(HighCapacityGenerator, self).__init__()\n", " # Enhanced architecture: Adding an extra hidden layer and increasing width\n", " self.fc = nn.Sequential(\n", " nn.Linear(input_dim, 128),\n", " nn.ReLU(),\n", " nn.Linear(128, 128),\n", " nn.ReLU(),\n", " nn.Linear(128, output_dim)\n", " )\n", "\n", " def forward(self, x):\n", " return self.fc(x)\n", "\n", "class Discriminator(nn.Module):\n", " def __init__(self, input_dim=2):\n", " super(Discriminator, self).__init__()\n", " self.fc = nn.Sequential(\n", " nn.Linear(input_dim, 50),\n", " nn.ReLU(),\n", " nn.Linear(50, 1),\n", " nn.Sigmoid()\n", " )\n", "\n", " def forward(self, x):\n", " return self.fc(x)\n", "\n", "# --- 2. Utility Functions ---\n", "\n", "def generate_real_data(batch_size=64, mean1=0.0, mean2=10.0, std=1.0):\n", " half_batch = batch_size // 2\n", " data1 = np.random.normal(mean1, std, (half_batch, 2))\n", " data2 = np.random.normal(mean2, std, (batch_size - half_batch, 2))\n", " combined_data = np.vstack([data1, data2])\n", " return torch.tensor(combined_data, dtype=torch.float32).to(device)\n", "\n", "def generate_noise(batch_size=64, noise_dim=10):\n", " return torch.randn(batch_size, noise_dim).to(device)\n", "\n", "# --- 3. Initialization ---\n", "\n", "noise_dim = 10\n", "generator = HighCapacityGenerator(input_dim=noise_dim).to(device)\n", "discriminator = Discriminator().to(device)\n", "\n", "criterion = nn.BCELoss()\n", "optimizer_g = optim.Adam(generator.parameters(), lr=0.001)\n", "optimizer_d = optim.Adam(discriminator.parameters(), lr=0.001)\n", "\n", "# Training parameters\n", "num_epochs = 10000\n", "batch_size = 64\n", "losses_d = []\n", "losses_g = []\n", "\n", "# --- 4. Training Loop ---\n", "\n", "print(f\"Training with High-Capacity Generator on {device}...\")\n", "\n", "for epoch in range(num_epochs):\n", " # --- Step A: Train Discriminator ---\n", " optimizer_d.zero_grad()\n", " \n", " # Real data labels = 1, Fake data labels = 0\n", " real_data = generate_real_data(batch_size)\n", " real_labels = torch.ones(batch_size, 1).to(device)\n", " \n", " noise = generate_noise(batch_size, noise_dim)\n", " fake_data = generator(noise).detach() # Detach to avoid updating Generator\n", " fake_labels = torch.zeros(batch_size, 1).to(device)\n", " \n", " # D Loss\n", " out_real = discriminator(real_data)\n", " loss_d_real = criterion(out_real, real_labels)\n", " \n", " out_fake = discriminator(fake_data)\n", " loss_d_fake = criterion(out_fake, fake_labels)\n", " \n", " loss_d = loss_d_real + loss_d_fake\n", " loss_d.backward()\n", " optimizer_d.step()\n", " \n", " # --- Step B: Train Generator ---\n", " optimizer_g.zero_grad()\n", " \n", " noise = generate_noise(batch_size, noise_dim)\n", " fake_data = generator(noise)\n", " \n", " # G wants D to think fake data is real (label = 1)\n", " out_fake_for_g = discriminator(fake_data)\n", " loss_g = criterion(out_fake_for_g, real_labels)\n", " \n", " loss_g.backward()\n", " optimizer_g.step()\n", " \n", " # Save statistics\n", " losses_d.append(loss_d.item())\n", " losses_g.append(loss_g.item())\n", "\n", " if epoch % 1000 == 0:\n", " print(f\"Epoch [{epoch}/{num_epochs}] | Loss D: {loss_d.item():.4f} | Loss G: {loss_g.item():.4f}\")\n", "\n", "# --- 5. Visualization ---\n", "\n", "# Plot Losses\n", "plt.figure(figsize=(12, 5))\n", "plt.subplot(1, 2, 1)\n", "plt.plot(losses_d, label='D Loss', color='red', alpha=0.5)\n", "plt.plot(losses_g, label='G Loss', color='blue', alpha=0.5)\n", "plt.title(\"Training Losses\")\n", "plt.legend()\n", "\n", "# Plot Generated Data vs Real Data\n", "plt.subplot(1, 2, 2)\n", "generator.eval()\n", "with torch.no_grad():\n", " test_noise = generate_noise(200, noise_dim)\n", " generated = generator(test_noise).cpu().numpy()\n", " real = generate_real_data(200).cpu().numpy()\n", "\n", "plt.scatter(real[:, 0], real[:, 1], c='red', label='Real Data', alpha=0.5)\n", "plt.scatter(generated[:, 0], generated[:, 1], c='blue', label='Generated Data', alpha=0.5)\n", "plt.title(\"High-Capacity Generator Results\")\n", "plt.legend()\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "86edb1c3", "metadata": {}, "source": [ "**Analysis of the Results:**\n", "\n", "Increasing Generator capacity without a corresponding increase in Discriminator capacity (or better loss functions) does not necessarily improve results. Here, the Generator became powerful enough to exploit the Discriminator's \"blind spots\" rather than learning the actual data distribution." ] }, { "cell_type": "markdown", "id": "eab6eb0c", "metadata": {}, "source": [ "#### 2. Implementation: High-Capacity Discriminator\n", "\n", "In this step, we investigate the effect of increasing the capacity of the Discriminator to match the complexity of the Generator. By providing the Discriminator with more layers and neurons, we aim to create a more sophisticated \"critic\" that can theoretically provide a more detailed loss landscape for the Generator to learn from.\n", "\n", "We modified the `Discriminator` class to include an additional hidden layer and increased the width of the layers from 50 to 128 neurons.\n", "\n", "**Original Discriminator:**\n", "```python\n", "class Discriminator(nn.Module):\n", " def __init__(self, input_dim=2):\n", " super(Discriminator, self).__init__()\n", " # Architecture: 2 -> 50 -> 1\n", " self.fc = nn.Sequential(\n", " nn.Linear(input_dim, 50),\n", " nn.ReLU(),\n", " nn.Linear(50, 1),\n", " nn.Sigmoid()\n", " )\n", "\n", "```\n", "**High-Capacity Discriminator:**\n", "```python\n", "class HighCapacityDiscriminator(nn.Module):\n", " def __init__(self, input_dim=2):\n", " super(HighCapacityDiscriminator, self).__init__()\n", " # Architecture: 2 -> 128 -> 128 -> 1\n", " self.fc = nn.Sequential(\n", " nn.Linear(input_dim, 128),\n", " nn.ReLU(),\n", " nn.Linear(128, 128),\n", " nn.ReLU(),\n", " # Increased capacity to match the Generator\n", " nn.Linear(128, 1),\n", " nn.Sigmoid()\n", " )\n", "```\n" ] }, { "cell_type": "code", "execution_count": 2, "id": "b08330e7", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training with High-Capacity Generator AND Discriminator on cpu...\n", "Epoch [0/10000] | Loss D: 1.3395 | Loss G: 0.6741\n", "Epoch [1000/10000] | Loss D: 0.9498 | Loss G: 1.0741\n", "Epoch [2000/10000] | Loss D: 1.8334 | Loss G: 1.0794\n", "Epoch [3000/10000] | Loss D: 1.4157 | Loss G: 0.7272\n", "Epoch [4000/10000] | Loss D: 1.2811 | Loss G: 0.7338\n", "Epoch [5000/10000] | Loss D: 1.2811 | Loss G: 0.8065\n", "Epoch [6000/10000] | Loss D: 1.2619 | Loss G: 0.9606\n", "Epoch [7000/10000] | Loss D: 1.1284 | Loss G: 0.9045\n", "Epoch [8000/10000] | Loss D: 1.2806 | Loss G: 0.8654\n", "Epoch [9000/10000] | Loss D: 1.1076 | Loss G: 0.8248\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import torch\n", "import torch.nn as nn\n", "import torch.optim as optim\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "\n", "# Check for GPU\n", "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", "\n", "# --- 1. Define High-Capacity Models ---\n", "\n", "class HighCapacityGenerator(nn.Module):\n", " def __init__(self, input_dim=10, output_dim=2):\n", " super(HighCapacityGenerator, self).__init__()\n", " # Architecture: 10 -> 128 -> 128 -> 2\n", " self.fc = nn.Sequential(\n", " nn.Linear(input_dim, 128),\n", " nn.ReLU(),\n", " nn.Linear(128, 128),\n", " nn.ReLU(),\n", " nn.Linear(128, output_dim)\n", " )\n", "\n", " def forward(self, x):\n", " return self.fc(x)\n", "\n", "class HighCapacityDiscriminator(nn.Module):\n", " def __init__(self, input_dim=2):\n", " super(HighCapacityDiscriminator, self).__init__()\n", " # Architecture: 2 -> 128 -> 128 -> 1\n", " # Increased capacity to match the Generator\n", " self.fc = nn.Sequential(\n", " nn.Linear(input_dim, 128),\n", " nn.ReLU(),\n", " nn.Linear(128, 128),\n", " nn.ReLU(),\n", " nn.Linear(128, 1),\n", " nn.Sigmoid()\n", " )\n", "\n", " def forward(self, x):\n", " return self.fc(x)\n", "\n", "# --- 2. Utility Functions ---\n", "\n", "def generate_real_data(batch_size=64, mean1=0.0, mean2=10.0, std=1.0):\n", " half_batch = batch_size // 2\n", " data1 = np.random.normal(mean1, std, (half_batch, 2))\n", " data2 = np.random.normal(mean2, std, (batch_size - half_batch, 2))\n", " combined_data = np.vstack([data1, data2])\n", " return torch.tensor(combined_data, dtype=torch.float32).to(device)\n", "\n", "def generate_noise(batch_size=64, noise_dim=10):\n", " return torch.randn(batch_size, noise_dim).to(device)\n", "\n", "# --- 3. Initialization ---\n", "\n", "noise_dim = 10\n", "generator = HighCapacityGenerator(input_dim=noise_dim).to(device)\n", "discriminator = HighCapacityDiscriminator().to(device)\n", "\n", "criterion = nn.BCELoss()\n", "optimizer_g = optim.Adam(generator.parameters(), lr=0.001)\n", "optimizer_d = optim.Adam(discriminator.parameters(), lr=0.001)\n", "\n", "# Training parameters\n", "num_epochs = 10000\n", "batch_size = 64\n", "losses_d = []\n", "losses_g = []\n", "\n", "# --- 4. Training Loop ---\n", "\n", "print(f\"Training with High-Capacity Generator AND Discriminator on {device}...\")\n", "\n", "for epoch in range(num_epochs):\n", " # --- Step A: Train Discriminator ---\n", " optimizer_d.zero_grad()\n", " \n", " real_data = generate_real_data(batch_size)\n", " real_labels = torch.ones(batch_size, 1).to(device)\n", " \n", " noise = generate_noise(batch_size, noise_dim)\n", " fake_data = generator(noise).detach() \n", " fake_labels = torch.zeros(batch_size, 1).to(device)\n", " \n", " # Forward passes\n", " out_real = discriminator(real_data)\n", " out_fake = discriminator(fake_data)\n", " \n", " loss_d = criterion(out_real, real_labels) + criterion(out_fake, fake_labels)\n", " loss_d.backward()\n", " optimizer_d.step()\n", " \n", " # --- Step B: Train Generator ---\n", " optimizer_g.zero_grad()\n", " \n", " noise = generate_noise(batch_size, noise_dim)\n", " fake_data = generator(noise)\n", " \n", " # Try to fool the high-capacity discriminator\n", " out_fake_for_g = discriminator(fake_data)\n", " loss_g = criterion(out_fake_for_g, real_labels)\n", " \n", " loss_g.backward()\n", " optimizer_g.step()\n", " \n", " losses_d.append(loss_d.item())\n", " losses_g.append(loss_g.item())\n", "\n", " if epoch % 1000 == 0:\n", " print(f\"Epoch [{epoch}/{num_epochs}] | Loss D: {loss_d.item():.4f} | Loss G: {loss_g.item():.4f}\")\n", "\n", "# --- 5. Visualization ---\n", "\n", "plt.figure(figsize=(12, 5))\n", "\n", "# Plot Losses\n", "plt.subplot(1, 2, 1)\n", "plt.plot(losses_d, label='D Loss (High Cap)', color='red', alpha=0.5)\n", "plt.plot(losses_g, label='G Loss (High Cap)', color='blue', alpha=0.5)\n", "plt.title(\"Balanced High-Capacity Training Losses\")\n", "plt.xlabel(\"Epochs\")\n", "plt.ylabel(\"Loss\")\n", "plt.legend()\n", "\n", "# Plot Results\n", "plt.subplot(1, 2, 2)\n", "generator.eval()\n", "with torch.no_grad():\n", " test_noise = generate_noise(300, noise_dim)\n", " generated = generator(test_noise).cpu().numpy()\n", " real = generate_real_data(300).cpu().numpy()\n", "\n", "plt.scatter(real[:, 0], real[:, 1], c='red', label='Real Data (2 Clusters)', alpha=0.4, s=15)\n", "plt.scatter(generated[:, 0], generated[:, 1], c='blue', label='Generated Data', alpha=0.6, s=15)\n", "plt.title(\"Results: Both Networks at High Capacity\")\n", "plt.legend()\n", "plt.tight_layout()\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "33389098", "metadata": {}, "source": [ "**Analysis of the Results:**\n", "\n", "Run the code multiple times and observe that the strategy changes every time; the Generator may focus on the top cluster, the bottom cluster, or a line between them. This occurs because the Generator discovers a small region that fools the Discriminator but fails to capture the true diversity of the data. Raw model power cannot overcome fundamental instabilities like oscillating losses and mode collapse without structural changes." ] }, { "cell_type": "markdown", "id": "76934816", "metadata": {}, "source": [ "#### 3. Implementation: Imbalance the Networks\n", "In this final experiment, we explore scenarios where one player in the adversarial game is significantly more powerful than the other. This helps visualize the \"delicate balancing act\" required for stable GAN training." ] }, { "cell_type": "code", "execution_count": 3, "id": "e8bc26ca", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Experiment 1: Powerful Generator vs. Weak Discriminator...\n", "Experiment 2: Weak Generator vs. Powerful Discriminator...\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import torch\n", "import torch.nn as nn\n", "import torch.optim as optim\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "\n", "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", "\n", "# --- 1. Define Architecture Variants ---\n", "\n", "class Generator(nn.Module):\n", " def __init__(self, hidden_dim):\n", " super().__init__()\n", " # hidden_dim=50 for original, 128 for high-capacity\n", " self.fc = nn.Sequential(\n", " nn.Linear(10, hidden_dim), nn.ReLU(),\n", " nn.Linear(hidden_dim, hidden_dim), nn.ReLU() if hidden_dim==128 else nn.Identity(),\n", " nn.Linear(hidden_dim, 2)\n", " )\n", " def forward(self, x): return self.fc(x)\n", "\n", "class Discriminator(nn.Module):\n", " def __init__(self, hidden_dim):\n", " super().__init__()\n", " # hidden_dim=50 for original, 128 for high-capacity\n", " self.fc = nn.Sequential(\n", " nn.Linear(2, hidden_dim), nn.ReLU(),\n", " nn.Linear(hidden_dim, hidden_dim), nn.ReLU() if hidden_dim==128 else nn.Identity(),\n", " nn.Linear(hidden_dim, 1), nn.Sigmoid()\n", " )\n", " def forward(self, x): return self.fc(x)\n", "\n", "# --- 2. Training Function ---\n", "\n", "def train_gan(gen, disc, title, epochs=5000):\n", " criterion = nn.BCELoss()\n", " opt_g = optim.Adam(gen.parameters(), lr=0.001)\n", " opt_d = optim.Adam(disc.parameters(), lr=0.001)\n", " \n", " for epoch in range(epochs):\n", " # Train Discriminator\n", " opt_d.zero_grad()\n", " real = generate_real_data(64)\n", " fake = gen(generate_noise(64)).detach()\n", " loss_d = criterion(disc(real), torch.ones(64, 1).to(device)) + \\\n", " criterion(disc(fake), torch.zeros(64, 1).to(device))\n", " loss_d.backward()\n", " opt_d.step()\n", "\n", " # Train Generator\n", " opt_g.zero_grad()\n", " fake = gen(generate_noise(64))\n", " loss_g = criterion(disc(fake), torch.ones(64, 1).to(device))\n", " loss_g.backward()\n", " opt_g.step()\n", " \n", " return gen\n", "\n", "# --- 3. Execution of Experiments ---\n", "\n", "print(\"Experiment 1: Powerful Generator vs. Weak Discriminator...\")\n", "# High-cap Gen (128) + Low-cap Disc (50)\n", "gen_1 = train_gan(Generator(128).to(device), Discriminator(50).to(device), \"G > D\")\n", "\n", "print(\"Experiment 2: Weak Generator vs. Powerful Discriminator...\")\n", "# Low-cap Gen (50) + High-cap Disc (128)\n", "gen_2 = train_gan(Generator(50).to(device), Discriminator(128).to(device), \"D > G\")\n", "\n", "# --- 4. Visualization ---\n", "\n", "plt.figure(figsize=(14, 6))\n", "\n", "def plot_result(gen, subplot_idx, title):\n", " plt.subplot(1, 2, subplot_idx)\n", " gen.eval()\n", " with torch.no_grad():\n", " fake = gen(generate_noise(300)).cpu().numpy()\n", " real = generate_real_data(300).cpu().numpy()\n", " plt.scatter(real[:, 0], real[:, 1], c='red', alpha=0.3, label='Real')\n", " plt.scatter(fake[:, 0], fake[:, 1], c='blue', alpha=0.6, label='Fake')\n", " plt.title(title)\n", " plt.legend()\n", "\n", "plot_result(gen_1, 1, \"Scenario A: Strong G (128) / Weak D (50)\")\n", "plot_result(gen_2, 2, \"Scenario B: Weak G (50) / Strong D (128)\")\n", "plt.tight_layout()\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "2038800f", "metadata": {}, "source": [ "**Scenario A: Strong Generator (128) / Weak Discriminator (50)**\n", "\n", "The Generator is upgraded to 128 hidden units with an extra layer, while the Discriminator remains at the original capacity of 50 units.\n", "\n", "Analysis of the Results: Increasing Generator capacity without a corresponding increase in Discriminator capacity does not necessarily improve results. The Generator became powerful enough to exploit the Discriminator's \"blind spots\" rather than learning the actual data distribution. Because the Discriminator is too weak to provide a complex boundary, the Generator easily \"wins\" by finding simple shortcuts.\n", "\n", "\n", "**Scenario B: Weak Generator (50) / Strong Discriminator (128)**\n", "\n", "The Generator remains at the original capacity of 50 units, while the Discriminator is upgraded to 128 hidden units with an extra layer.\n", "\n", "Analysis of the Results: When the Discriminator is significantly more powerful, it becomes too effective at distinguishing real from fake data. The Generator often gets stuck in severe mode collapse, repeatedly attempting to cover only one small region." ] }, { "cell_type": "markdown", "id": "ae77d4b1", "metadata": {}, "source": [ "### Solutions: Exercise 1 Follow-up Questions\n", "\n", "#### 1) Observation of Final Generated Data and Loss Curves\n", "* **High-Capacity Generator / Low-Capacity Discriminator**: The generator has enough capacity to model both data modes, but because the discriminator is weak, it may not provide a sharp enough signal to force the generator to cover both clusters accurately. The loss curves often show the generator \"winning\" easily, potentially leading to **mode collapse** where it only covers one cluster.\n", "* **Low-Capacity Generator / High-Capacity Discriminator**: The discriminator becomes too powerful for the generator to fool. This often results in a **\"vanishing gradient\"** problem where the generator’s loss becomes very high and stays flat because it cannot find any way to improve against the expert discriminator. The resulting data plot typically shows poor coverage of the real data distribution.\n", "* **Balanced High-Capacity (Both 128 units)**: The loss curves show intense oscillations as both networks improve together. The final data plot shows the best attempt at covering both clusters, as both agents are matched in their ability to learn the complex multimodal distribution.\n", "\n", "#### 2) Best Combination for Covering Both Data Modes\n", "The **balanced high-capacity combination** (where both the Generator and Discriminator have 128-unit hidden layers) produces the best results. This setup allows the generator to capture the complexity of the two distinct clusters (modes) while the discriminator is strong enough to provide the necessary feedback to ensure the generator doesn't just settle for one \"easy\" cluster.\n", "\n", "#### 3) Impact of Imbalanced Power\n", "* **What happens**: When one network is significantly more powerful, the competitive nature of the GAN breaks down.\n", " * If the **Discriminator is too strong**, the Generator receives no useful gradient information to learn from and essentially gets \"stuck\".\n", " * If the **Generator is too strong**, it easily finds a single point that fools the weak Discriminator and stops exploring other parts of the data distribution (**mode collapse**).\n", "* **The \"Delicate Balancing Act\"**: This relates to the game-theoretic foundation of GANs, where the goal is to reach a **Nash equilibrium**. Training only succeeds if both players are matched; if the competition isn't fair, one side dominates, the minimax game ends prematurely, and the generator never learns to replicate the true diversity of the real data distribution." ] }, { "cell_type": "markdown", "id": "4310f9bb", "metadata": {}, "source": [ "### Exercise 2: Implementing the Wasserstein GAN (WGAN)\n", "\n", "The Wasserstein GAN (WGAN) addresses the fundamental training instabilities of the Vanilla GAN by replacing the classification-based loss with a metric based on the **Earth Mover's distance**. This modification provides more stable gradients and helps mitigate mode collapse by providing a continuous signal even when distributions are disjoint.\n", "\n", "#### 2.1. Mathematical Objective and Constraints \n", "Unlike the original formulation that uses cross-entropy loss, the WGAN seeks to minimize the Wasserstein-1 distance. This is achieved through the following architectural and game-theoretic changes: \n", "* **Discriminator (Critic)**: In WGAN, the discriminator $D$ is called a **Critic** because it outputs a scalar score representing \"realness\" rather than a probability. \n", "* **Lipschitz Constraint**: The [WGAN paper](https://arxiv.org/abs/1701.07875) showed that the Critic must be $1$-Lipschitz continuous. This means that the Critic must have a maximum gradient, which makes GAN training significantly more stable. \n", " * **Providing Information**: A \"perfect\" traditional discriminator provides nearly zero gradient information when it successfully rejects fake samples. By limiting the gradient (Lipschitz constraint), the Critic is forced to be a \"worse\" binary classifier but a much better teacher, providing a non-zero gradient even in disjoint regions. \n", " * **Visualization**: This is visible in Figure 2 of the [WGAN paper](https://arxiv.org/abs/1701.07875), where a standard discriminator's gradient saturates at nearly zero, while the WGAN critic maintains a non-zero gradient effective for training. \n", " * **Weight Clipping**: The WGAN enforces the Lipschitz constraint by clipping the Critic's weights (e.g., to $[-0.01, 0.01]$) after every update.\n", "\n", "**Implementation: WGAN with Weight Clipping:**\n", "\n", "To implement the WGAN, the following modifications were made to the architecture:\n", "\n", "* **Removal of Sigmoid**: The `nn.Sigmoid()` activation was removed from the Critic to allow for unbounded real scores.\n", "* **Wasserstein Objective**: The `nn.BCELoss()` was replaced with the mean difference objective.\n", "* **Optimizer Choice**: Switched from `Adam` to `RMSprop` for improved training stability.\n", "\n", "```python\n", "# 1. Critic Architecture - Unbounded scores\n", "class Critic(nn.Module):\n", " def __init__(self, input_dim=2):\n", " super(Critic, self).__init__()\n", " self.fc = nn.Sequential(\n", " nn.Linear(input_dim, 50),\n", " nn.ReLU(),\n", " nn.Linear(50, 1) # No Sigmoid for WGAN\n", " )\n", "\n", " def forward(self, x):\n", " return self.fc(x)\n", "\n", "# 2. Training Loop Logic (WGAN Modification)\n", "# --- Train Critic ---\n", "optimizer_d.zero_grad()\n", "# Critic aims to maximize (mean(D(real)) - mean(D(fake)))\n", "# In PyTorch, we minimize the negative of that value\n", "loss_d = -torch.mean(critic(real_data)) + torch.mean(critic(fake_data))\n", "loss_d.backward()\n", "optimizer_d.step()\n", "\n", "# Enforce Lipschitz Constraint via Weight Clipping\n", "for p in Critic.parameters():\n", " p.data.clamp_(-0.01, 0.01)\n", "\n", "# --- Train Generator ---\n", "optimizer_g.zero_grad()\n", "# G aims to maximize mean(D(fake))\n", "loss_g = -torch.mean(critic(fake_data))\n", "loss_g.backward()\n", "optimizer_g.step()\n", "```" ] }, { "cell_type": "code", "execution_count": 4, "id": "8ecf57b4", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training Wasserstein GAN (WGAN) on cpu...\n", "Epoch [0/10000] | Critic Loss: -0.0298 | Gen Loss: -0.0096\n", "Epoch [1000/10000] | Critic Loss: -0.0230 | Gen Loss: -0.0112\n", "Epoch [2000/10000] | Critic Loss: -0.0169 | Gen Loss: -0.0166\n", "Epoch [3000/10000] | Critic Loss: -0.0088 | Gen Loss: -0.0237\n", "Epoch [4000/10000] | Critic Loss: -0.0019 | Gen Loss: -0.0207\n", "Epoch [5000/10000] | Critic Loss: -0.0014 | Gen Loss: -0.0070\n", "Epoch [6000/10000] | Critic Loss: -0.0013 | Gen Loss: -0.0071\n", "Epoch [7000/10000] | Critic Loss: -0.0014 | Gen Loss: -0.0065\n", "Epoch [8000/10000] | Critic Loss: -0.0014 | Gen Loss: -0.0053\n", "Epoch [9000/10000] | Critic Loss: -0.0013 | Gen Loss: -0.0056\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import torch\n", "import torch.nn as nn\n", "import torch.optim as optim\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "\n", "# Check for GPU\n", "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", "\n", "# --- 1. Define the Models ---\n", "\n", "class Generator(nn.Module):\n", " def __init__(self, input_dim=10, output_dim=2):\n", " super(Generator, self).__init__()\n", " self.fc = nn.Sequential(\n", " nn.Linear(input_dim, 50),\n", " nn.ReLU(),\n", " nn.Linear(50, 2)\n", " )\n", "\n", " def forward(self, x):\n", " return self.fc(x)\n", "\n", "# WGAN Modification: The Discriminator is now a \"Critic\"\n", "class Critic(nn.Module):\n", " def __init__(self, input_dim=2):\n", " super(Critic, self).__init__()\n", " self.fc = nn.Sequential(\n", " nn.Linear(input_dim, 50),\n", " nn.ReLU(),\n", " nn.Linear(50, 1) \n", " # Note: nn.Sigmoid() removed for WGAN to allow unbounded real scores\n", " )\n", "\n", " def forward(self, x):\n", " return self.fc(x)\n", "\n", "# --- 2. Utility Functions ---\n", "\n", "def generate_real_data(batch_size=64, mean1=0.0, mean2=10.0, std=1.0):\n", " half_batch = batch_size // 2\n", " data1 = np.random.normal(mean1, std, (half_batch, 2))\n", " data2 = np.random.normal(mean2, std, (batch_size - half_batch, 2))\n", " combined_data = np.vstack([data1, data2])\n", " return torch.tensor(combined_data, dtype=torch.float32).to(device)\n", "\n", "def generate_noise(batch_size=64, noise_dim=10):\n", " return torch.randn(batch_size, noise_dim).to(device)\n", "\n", "# --- 3. Initialization ---\n", "\n", "noise_dim = 10\n", "generator = Generator(input_dim=noise_dim).to(device)\n", "critic = Critic().to(device)\n", "\n", "# WGAN Modification: Use RMSprop instead of Adam for improved stability\n", "optimizer_g = optim.RMSprop(generator.parameters(), lr=0.00005)\n", "optimizer_d = optim.RMSprop(critic.parameters(), lr=0.00005)\n", "\n", "# Training parameters\n", "num_epochs = 10000\n", "batch_size = 64\n", "losses_d = []\n", "losses_g = []\n", "\n", "# --- 4. Training Loop ---\n", "\n", "print(f\"Training Wasserstein GAN (WGAN) on {device}...\")\n", "\n", "for epoch in range(num_epochs):\n", " # --- Step A: Train Critic (Discriminator) ---\n", " optimizer_d.zero_grad()\n", " \n", " real_data = generate_real_data(batch_size)\n", " noise = generate_noise(batch_size, noise_dim)\n", " fake_data = generator(noise).detach() \n", " \n", " # WGAN Loss: -torch.mean(D(real)) + torch.mean(D(fake))\n", " # This maximizes the difference between the scores of real and fake data\n", " loss_d = -torch.mean(critic(real_data)) + torch.mean(critic(fake_data))\n", " \n", " loss_d.backward()\n", " optimizer_d.step()\n", " \n", " # WGAN Modification: Weight Clipping to enforce the Lipschitz constraint\n", " for p in critic.parameters():\n", " p.data.clamp_(-0.01, 0.01)\n", " \n", " # --- Step B: Train Generator ---\n", " optimizer_g.zero_grad()\n", " \n", " noise = generate_noise(batch_size, noise_dim)\n", " fake_data = generator(noise)\n", " \n", " # G wants to maximize the Critic's score for fake data: -torch.mean(D(fake))\n", " loss_g = -torch.mean(critic(fake_data))\n", " \n", " loss_g.backward()\n", " optimizer_g.step()\n", " \n", " # Save statistics\n", " losses_d.append(loss_d.item())\n", " losses_g.append(loss_g.item())\n", "\n", " if epoch % 1000 == 0:\n", " print(f\"Epoch [{epoch}/{num_epochs}] | Critic Loss: {loss_d.item():.4f} | Gen Loss: {loss_g.item():.4f}\")\n", "\n", "# --- 5. Visualization ---\n", "\n", "plt.figure(figsize=(12, 5))\n", "\n", "# Plot Losses\n", "plt.subplot(1, 2, 1)\n", "plt.plot(losses_d, label='Critic Loss', color='red', alpha=0.5)\n", "plt.plot(losses_g, label='Gen Loss', color='blue', alpha=0.5)\n", "plt.title(\"WGAN Training Losses\")\n", "plt.xlabel(\"Epochs\")\n", "plt.legend()\n", "\n", "# Plot Results\n", "plt.subplot(1, 2, 2)\n", "generator.eval()\n", "with torch.no_grad():\n", " test_noise = generate_noise(300, noise_dim)\n", " generated = generator(test_noise).cpu().numpy()\n", " real = generate_real_data(300).cpu().numpy()\n", "\n", "plt.scatter(real[:, 0], real[:, 1], c='red', label='Real Data', alpha=0.4)\n", "plt.scatter(generated[:, 0], generated[:, 1], c='blue', label='Generated Data', alpha=0.6)\n", "plt.title(\"WGAN Results\")\n", "plt.legend()\n", "plt.tight_layout()\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "a9d9223c", "metadata": {}, "source": [ "The distinct behaviors of the Vanilla GAN and Wasserstein GAN Convergence originate from the objectives of their respective loss functions. In Exercise 1, the Vanilla GAN minimizes the [Jensen-Shannon Divergence](https://en.wikipedia.org/wiki/Jensen%E2%80%93Shannon_divergence). When the generated data falls in the 'dead zone' (disjoint distributions), gradients vanish because the JSD creates a constant loss with no directional signal. To ensure some signal, the generator often collapses onto a single cluster (Mode Collapse) rather than traversing the gap. In contrast, the WGAN in Exercise 2 optimizes the Wasserstein (Earth Mover's) distance, a geometric metric that measures the physical cost of transporting mass to the target locations. Because this loss is proportional to distance rather than binary correctness, the model is not strictly punished for occupying the empty space; instead, it minimizes the total transport cost by forming a \"bridge\" across the gap, effectively averaging the two clusters rather than committing to just one." ] }, { "cell_type": "markdown", "id": "796fc26e", "metadata": {}, "source": [ "## Solutions for Tutorial 3.2: Evading ML-based IDS\n", "\n", "### Exercise 1: Attack Success vs. Perturbation Strength\n", "The goal of this exercise is to adjust the **Perturbation Penalty ($\\lambda_{pert}$)** and the generator's learning rate to observe how the modifications affect the ability to evade the target IDS. This hyperparameter defines the critical trade-off between the effectiveness of the attack and its subtlety. \n", "\n", "#### Implementation: AdvGAN Tuning \n", "We modified the training process to observe how the **$\\mathcal{L}_{pert}$** component influences the final detection rate of the victim IDS. This loss penalizes the perturbation size, ensuring that the adversarial changes remain small and subtle.\n", "\n", "* **High-Stealth Configuration ($\\lambda_{pert}=10.0$):** In this scenario, the generator is strictly penalized for making large changes. The objective is to keep the adversarial traffic as close as possible to the original malicious distribution to avoid detection by both the IDS and any potential secondary realism checks.\n", "\n", "* **Aggressive Configuration ($\\lambda_{pert}=0.01$):** In this scenario, the penalty is significantly reduced, allowing the generator to apply larger perturbations. This gives the model more freedom to reach the target IDS's \"blind spots\" and achieve a higher rate of evasion. " ] }, { "cell_type": "code", "execution_count": 5, "id": "b08d035d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "šŸš€ Training AdvGAN: lambda_pert=10.0, lr_g=0.001\n", "āœ… Final Detection Rate: 73.86%\n", "\n", "šŸš€ Training AdvGAN: lambda_pert=0.01, lr_g=0.001\n", "āœ… Final Detection Rate: 62.94%\n" ] }, { "data": { "text/plain": [ "AdvGAN_Generator(\n", " (model): Sequential(\n", " (0): Linear(in_features=77, out_features=128, bias=True)\n", " (1): ReLU()\n", " (2): Linear(in_features=128, out_features=128, bias=True)\n", " (3): ReLU()\n", " (4): Linear(in_features=128, out_features=77, bias=True)\n", " (5): Tanh()\n", " )\n", ")" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import os\n", "import pandas as pd\n", "import numpy as np\n", "import torch\n", "import torch.nn as nn\n", "import torch.optim as optim\n", "from torch.utils.data import TensorDataset, DataLoader\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.preprocessing import StandardScaler\n", "from sklearn.ensemble import RandomForestClassifier\n", "from sklearn.metrics import accuracy_score\n", "\n", "# --- 1. CONFIGURATION & DATA LOADING ---\n", "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", "DATA_DIR = \"./cicids2017\"\n", "data_files = [\n", " 'Benign-Monday-no-metadata.parquet', 'Bruteforce-Tuesday-no-metadata.parquet',\n", " 'Portscan-Friday-no-metadata.parquet', 'WebAttacks-Thursday-no-metadata.parquet',\n", " 'DoS-Wednesday-no-metadata.parquet', 'DDoS-Friday-no-metadata.parquet',\n", " 'Infiltration-Thursday-no-metadata.parquet', 'Botnet-Friday-no-metadata.parquet'\n", "]\n", "\n", "# Merge and Clean\n", "df_list = [pd.read_parquet(os.path.join(DATA_DIR, f)) for f in data_files]\n", "df_data = pd.concat(df_list, axis=0, ignore_index=True).drop_duplicates()\n", "df_numeric = df_data.select_dtypes(include=[np.number]).replace([np.inf, -np.inf], np.nan).dropna()\n", "\n", "X = df_numeric.drop(columns='Label', errors='ignore')\n", "y = df_data.loc[df_numeric.index, \"Label\"].map({'Benign': 0}).fillna(1).astype(int)\n", "\n", "# Scale and Split\n", "scaler = StandardScaler()\n", "X_scaled = scaler.fit_transform(X)\n", "X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.2, random_state=42, stratify=y)\n", "input_dim = X_train.shape[1]\n", "\n", "# --- 2. VICTIM IDS MODEL (KDDNet) ---\n", "class KDDNet(nn.Module):\n", " def __init__(self, input_dim):\n", " super(KDDNet, self).__init__()\n", " self.model = nn.Sequential(\n", " nn.Linear(input_dim, 128), nn.Tanh(), nn.Dropout(0.3),\n", " nn.Linear(128, 64), nn.Tanh(), nn.Dropout(0.3),\n", " nn.Linear(64, 1), nn.Sigmoid()\n", " )\n", " def forward(self, x): return self.model(x)\n", "\n", "target_ids_pytorch = KDDNet(input_dim).to(device)\n", "train_loader = DataLoader(TensorDataset(torch.tensor(X_train, dtype=torch.float32), \n", " torch.tensor(y_train.values, dtype=torch.float32).unsqueeze(1)), \n", " batch_size=512, shuffle=True)\n", "\n", "# Quick training of the victim\n", "target_opt = optim.Adam(target_ids_pytorch.parameters(), lr=0.001)\n", "criterion_bce = nn.BCELoss()\n", "target_ids_pytorch.train()\n", "for _ in range(3):\n", " for X_b, y_b in train_loader:\n", " X_b, y_b = X_b.to(device), y_b.to(device)\n", " target_opt.zero_grad()\n", " criterion_bce(target_ids_pytorch(X_b), y_b).backward()\n", " target_opt.step()\n", "\n", "# --- 3. ADVGAN ARCHITECTURE ---\n", "class AdvGAN_Generator(nn.Module):\n", " def __init__(self, input_dim):\n", " super().__init__()\n", " self.model = nn.Sequential(nn.Linear(input_dim, 128), nn.ReLU(), nn.Linear(128, 128), nn.ReLU(),\n", " nn.Linear(128, input_dim), nn.Tanh())\n", " def forward(self, x, mask):\n", " return x + (self.model(x) * 0.1) * mask\n", "\n", "class AdvGAN_Discriminator(nn.Module):\n", " def __init__(self, input_dim):\n", " super().__init__()\n", " self.model = nn.Sequential(nn.Linear(input_dim, 128), nn.ReLU(), nn.Linear(128, 1), nn.Sigmoid())\n", " def forward(self, x): return self.model(x)\n", "\n", "# Setup Masks\n", "feature_names = X.columns.tolist()\n", "immutable_features = ['Protocol', 'Fwd PSH Flags', 'FIN Flag Count', 'SYN Flag Count']\n", "mask = np.ones(len(feature_names))\n", "immutable_indices = [feature_names.index(f) for f in immutable_features if f in feature_names]\n", "mask[immutable_indices] = 0\n", "mask_tensor = torch.tensor(mask, dtype=torch.float32, device=device)\n", "\n", "# --- 4. EXERCISE 1: TRAINING FUNCTION WITH RETURN ---\n", "def train_advgan_tuned(lambda_pert_val, lr_g):\n", " print(f\"\\nšŸš€ Training AdvGAN: lambda_pert={lambda_pert_val}, lr_g={lr_g}\")\n", " gen_model = AdvGAN_Generator(input_dim).to(device)\n", " disc_model = AdvGAN_Discriminator(input_dim).to(device)\n", " opt_g = optim.Adam(gen_model.parameters(), lr=lr_g)\n", " opt_d = optim.Adam(disc_model.parameters(), lr=0.001)\n", " loss_mse = nn.MSELoss()\n", " \n", " mal_l = DataLoader(TensorDataset(torch.tensor(X_train[y_train == 1], dtype=torch.float32)), batch_size=512, shuffle=True)\n", " ben_l = DataLoader(TensorDataset(torch.tensor(X_train[y_train == 0], dtype=torch.float32)), batch_size=512, shuffle=True)\n", "\n", " for epoch in range(3):\n", " for (mal_b,), (ben_b,) in zip(mal_l, ben_l):\n", " mal_b, ben_b = mal_b.to(device), ben_b.to(device)\n", " # Disc update\n", " opt_d.zero_grad()\n", " d_loss = (criterion_bce(disc_model(ben_b), torch.ones(ben_b.size(0), 1).to(device)) + \n", " criterion_bce(disc_model(gen_model(mal_b, mask_tensor).detach()), torch.zeros(mal_b.size(0), 1).to(device))) / 2\n", " d_loss.backward(); opt_d.step()\n", " # Gen update\n", " opt_g.zero_grad()\n", " adv_g = gen_model(mal_b, mask_tensor)\n", " g_loss = criterion_bce(disc_model(adv_g), torch.ones(mal_b.size(0), 1).to(device)) + \\\n", " 1.0 * criterion_bce(target_ids_pytorch(adv_g), torch.zeros(mal_b.size(0), 1).to(device)) + \\\n", " lambda_pert_val * loss_mse(adv_g, mal_b)\n", " g_loss.backward(); opt_g.step()\n", "\n", " # Evaluation\n", " gen_model.eval()\n", " X_test_mal_t = torch.tensor(X_test[y_test == 1], dtype=torch.float32).to(device)\n", " with torch.no_grad():\n", " preds = (target_ids_pytorch(gen_model(X_test_mal_t, mask_tensor)).squeeze() > 0.5)\n", " print(f\"āœ… Final Detection Rate: {preds.float().mean().item():.2%}\")\n", " \n", " return gen_model # CRITICAL: Return the model to avoid TypeError\n", "\n", "# Execute Exercise 1\n", "train_advgan_tuned(lambda_pert_val=10.0, lr_g=0.001)\n", "train_advgan_tuned(lambda_pert_val=0.01, lr_g=0.001)\n", "\n" ] }, { "cell_type": "markdown", "id": "26d1c889", "metadata": {}, "source": [ "**Analysis of the Results:**\n", "Based on the execution output, we observe a clear correlation between perturbation strength and evasion success: \n", "* **Observation of Results**: With $\\lambda_{pert}=10.0$ (High Stealth), the final detection rate was higher than when the penalty was lowered to $0.01$ (Aggressive). \n", "* **Evasion Effectiveness**: The results confirm that reducing the perturbation penalty allows the generator to craft more effective adversarial samples. The lower detection rate in the aggressive configuration indicates that larger modifications allow the traffic to move further away from the malicious cluster learned by the IDS. " ] }, { "cell_type": "markdown", "id": "51556a86", "metadata": {}, "source": [ "### Exercise 2: Compare Models\n", "\n", "The objective of this exercise is to evaluate the **transferability** of adversarial attacks. Transferability refers to the phenomenon in which adversarial examples crafted to fool one specific model (e.g., a Neural Network) are also effective at evading other architectures (e.g., a Random Forest) trained on the same data. \n", "\n", "#### Implementation: Testing Transferability \n", "In this step, we train a **Random Forest (RF)** classifier to serve as an alternative IDS. We then test the adversarial samples generated by the **AdvGAN** to see whether they can also evade the Random Forest. " ] }, { "cell_type": "code", "execution_count": 6, "id": "cc45d27e", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "šŸš€ Training AdvGAN: lambda_pert=0.01, lr_g=0.001\n", "āœ… Final Detection Rate: 76.87%\n", "\n", "🌲 Training Random Forest as Alternative IDS...\n", "šŸ“Š RF Detection Rate on Transferred Attacks: 0.00%\n" ] } ], "source": [ "# --- 5. EXECUTION ---\n", "# Execute Exercise 1\n", "generator = train_advgan_tuned(lambda_pert_val=0.01, lr_g=0.001)\n", "\n", "# --- 6. EXERCISE 2: TRANSFERABILITY ---\n", "print(\"\\n🌲 Training Random Forest as Alternative IDS...\")\n", "rf_ids = RandomForestClassifier(n_estimators=50, max_depth=10, random_state=42, n_jobs=-1)\n", "rf_ids.fit(X_train, y_train)\n", "\n", "with torch.no_grad():\n", " X_test_mal_t = torch.tensor(X_test[y_test == 1], dtype=torch.float32).to(device)\n", " adv_test_samples = generator(X_test_mal_t, mask_tensor).cpu().numpy()\n", "\n", "rf_adv_det_rate = accuracy_score(y_test[y_test == 1], rf_ids.predict(adv_test_samples))\n", "print(f\"šŸ“Š RF Detection Rate on Transferred Attacks: {rf_adv_det_rate:.2%}\")" ] }, { "cell_type": "markdown", "id": "cfaf7d81", "metadata": {}, "source": [ "**Analysis of the Results:**\n", "The execution results reveal significant security implications regarding the robustness of ML-based systems:\n", "\n", "* **Black-Box Evasion Potential**: This confirms that GAN-based attacks are highly effective in **black-box scenarios**, where the attacker has no internal model knowledge and only observes outputs. An attacker can successfully evade a target IDS by training a generator against a substitute model (the Neural Network), which then generates samples that fool a completely different architecture (the Random Forest).\n", "* **Fragility of Decision Boundaries**: The total failure of the Random Forest highlights the fragility of data-driven IDS boundaries. These models often rely on non-causal statistical correlations rather than a proper understanding of threats, allowing carefully crafted adversarial attacks to push samples across decision boundaries." ] }, { "cell_type": "markdown", "id": "0a827991", "metadata": {}, "source": [ "---\n", "\n", "[![Star our repository](https://img.shields.io/static/v1.svg?logo=star&label=⭐&message=Star%20Our%20Repository&color=yellow)](https://github.com/clandolt/mlcysec_notebooks/) If you found this tutorial helpful, please **⭐ star our repository** to show your support. \n", "[![Ask questions](https://img.shields.io/static/v1.svg?logo=star&label=ā”&message=Ask%20Questions&color=9cf)](https://github.com/clandolt/mlcysec_notebooks/issues) For any **questions**, **typos**, or **bugs**, kindly open an issue on GitHub — we appreciate your feedback!\n", "\n", "---" ] }, { "cell_type": "markdown", "id": "7e90861a", "metadata": {}, "source": [] } ], "metadata": { "kernelspec": { "display_name": "mlcysec25", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.2" } }, "nbformat": 4, "nbformat_minor": 5 }