# What Is the Learning Path for AI For Beginners? A 12-Week Curriculum Guide

> Start your AI journey with our 12-week learning path. Master AI fundamentals, neural networks, computer vision, NLP, and ethics with hands-on PyTorch and TensorFlow.

- Repository: [Microsoft/AI-For-Beginners](https://github.com/microsoft/AI-For-Beginners)
- Tags: getting-started
- Published: 2026-08-19

---

**The AI For Beginners learning path is a linear, 12-week curriculum that takes newcomers from foundational AI concepts through neural networks, computer vision, natural language processing, ethics, and multi-modal systems using hands-on PyTorch and TensorFlow notebooks.**

The `microsoft/AI-For-Beginners` repository provides a structured, beginner-friendly roadmap that defines the complete learning path for AI For Beginners. According to the source code, the full syllabus is mapped in the repository's main [`README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md) (lines 81–118), which links every lesson to executable Jupyter notebooks and optional labs. This guide breaks down the exact phases, files, and code samples you will encounter.

## The 12-Week AI For Beginners Curriculum Roadmap

The curriculum is divided into discrete phases that progress from environment setup to advanced applications. Each phase maps to a specific directory under the `lessons/` folder:

- **Phase 0 – Course Setup**: [`lessons/0-course-setup/setup.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/0-course-setup/setup.md) covers conda environment creation, VS Code configuration, and dev-container setup.
- **Phase I – Introduction to AI**: [`lessons/1-Intro/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/1-Intro/README.md) presents the history and high-level view of artificial intelligence.
- **Phase II – Symbolic AI**: [`lessons/2-Symbolic/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/2-Symbolic/README.md) explores knowledge representation, expert systems, and ontologies.
- **Phase III – Neural Networks (Foundations)**: [`lessons/3-NeuralNetworks/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/3-NeuralNetworks/README.md) introduces the perceptron, multilayer perceptron, over-fitting, and framework basics for PyTorch and TensorFlow.
- **Phase IV – Computer Vision**: [`lessons/4-ComputerVision/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/4-ComputerVision/README.md) teaches OpenCV fundamentals, convolutional neural networks, transfer learning, GANs, segmentation, and object detection.
- **Phase V – Natural Language Processing**: [`lessons/5-NLP/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/5-NLP/README.md) covers text vectorisation (Bag-of-Words, TF-IDF), word embeddings (Word2Vec, GloVe), RNNs, transformers, named entity recognition, and large-language-model prompting.
- **Phase VI – Other AI Techniques**: [`lessons/6-Other/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/6-Other/README.md) surveys genetic algorithms, deep reinforcement learning, and multi-agent systems.
- **Phase VII – AI Ethics**: [`lessons/7-Ethics/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/7-Ethics/README.md) focuses on responsible AI principles, bias, transparency, and privacy.
- **Phase IX – Extras**: [`lessons/X-Extras/X1-MultiModal/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/X-Extras/X1-MultiModal/README.md) dives into multi-modal networks such as CLIP and VQ-GAN.
- **Hands-On Examples**: [`examples/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/examples/README.md) collects quick-start notebooks including Hello AI World, a simple neural network, an image classifier, and sentiment analysis.

This table-of-contents structure makes the learning path for AI For Beginners intentionally modular: you can follow it linearly or jump to a specific domain while relying on earlier sections for prerequisites.

## How Each Lesson Is Structured

Every lesson in the repository follows a consistent three-part format to reinforce theory with practice:

1. **Pre-reading material** – Short theory or background reading that introduces the core concepts.
2. **Executable Jupyter notebooks** – Code samples provided for **PyTorch** and **TensorFlow** (or both), allowing learners to run implementations directly.
3. **Labs (optional)** – Deeper, practice-focused tasks that reinforce the chapter's concepts.

The repository also links to relevant **Microsoft Learn** modules for extended reading, and includes a Vue.js-based quiz application in [`etc/quiz-app/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/etc/quiz-app/README.md) to test comprehension.

## Hands-On Code Examples from the Curriculum

The AI For Beginners learning path emphasizes a "learn-by-doing" philosophy. Below are minimal, runnable snippets taken directly from the repository's example notebooks.

### Hello AI World (NumPy)

This NumPy-only linear classifier appears in the **Hello AI World** example notebook ([`examples/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/examples/README.md), lines 31–37):

```python
import numpy as np

# Simple linear classifier on a tiny synthetic dataset

X = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([0, 0, 0, 1])               # AND logic

weights = np.random.rand(2)
bias = np.random.rand()

def predict(x):
    return int(np.dot(weights, x) + bias > 0.5)

print("Predictions:", [predict(x) for x in X])

```

### Simple Neural Network (PyTorch)

This one-hidden-layer network for binary classification is taken from the **Simple Neural Network** notebook in the examples folder:

```python
import torch
import torch.nn as nn
import torch.optim as optim

# One-hidden-layer network for binary classification

model = nn.Sequential(
    nn.Linear(2, 4),
    nn.ReLU(),
    nn.Linear(4, 1),
    nn.Sigmoid()
)

criterion = nn.BCELoss()
optimizer = optim.Adam(model.parameters(), lr=0.01)

X = torch.tensor([[0.,0.],[0.,1.],[1.,0.],[1.,1.]])
y = torch.tensor([[0.],[0.],[0.],[1.]])

for epoch in range(500):
    pred = model(X)
    loss = criterion(pred, y)
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()

```

### Image Classification with TensorFlow

Found in the **Computer Vision** lesson ([`lessons/4-ComputerVision/07-ConvNets/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/4-ComputerVision/07-ConvNets/README.md), lines 96–99), this snippet builds a convolutional network on CIFAR-10:

```python
import tensorflow as tf
from tensorflow.keras import layers, models
import matplotlib.pyplot as plt

# Load CIFAR-10 dataset (built-in)

(x_train, y_train), (x_test, y_test) = tf.keras.datasets.cifar10.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

model = models.Sequential([
    layers.Conv2D(32, (3,3), activation='relu', input_shape=(32,32,3)),
    layers.MaxPooling2D((2,2)),
    layers.Flatten(),
    layers.Dense(64, activation='relu'),
    layers.Dense(10, activation='softmax')
])

model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

model.fit(x_train, y_train, epochs=5, validation_split=0.1)

```

### Text Sentiment Analysis (TensorFlow)

This embedding-based classifier appears in the **Text Sentiment** example notebook ([`examples/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/examples/README.md)):

```python
import tensorflow as tf
from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.preprocessing.sequence import pad_sequences

sentences = ["I love this product!", "This is the worst experience."]
labels = [1, 0]

tokenizer = Tokenizer(num_words=1000, oov_token="<OOV>")
tokenizer.fit_on_texts(sentences)
sequences = tokenizer.texts_to_sequences(sentences)
padded = pad_sequences(sequences, maxlen=10, padding='post')

model = tf.keras.Sequential([
    tf.keras.layers.Embedding(1000, 16, input_length=10),
    tf.keras.layers.GlobalAveragePooling1D(),
    tf.keras.layers.Dense(16, activation='relu'),
    tf.keras.layers.Dense(1, activation='sigmoid')
])

model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
model.fit(padded, labels, epochs=10)

```

## Summary

- The **learning path for AI For Beginners** is a 12-week, linear curriculum mapped in [`README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/README.md) (lines 81–118).
- It covers setup, symbolic AI, neural networks, computer vision, NLP, other techniques, ethics, and multi-modal extras.
- Every lesson pairs **pre-reading** with **executable PyTorch and TensorFlow notebooks** and optional labs.
- Quick-start examples in [`examples/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/examples/README.md) let learners run NumPy, PyTorch, and TensorFlow code immediately.
- The repository is modular, so you can study sequentially or target specific domains while using earlier lessons for prerequisites.

## Frequently Asked Questions

### How long does it take to complete the AI For Beginners learning path?

The curriculum is designed as a 12-week program, with each phase roughly corresponding to one week of study. However, because the repository is modular, learners can adjust the pace or focus on specific topics such as computer vision or NLP without completing every preceding lab.

### What frameworks does the AI For Beginners curriculum use?

The course primarily uses **PyTorch** and **TensorFlow**. Many lessons provide parallel Jupyter notebooks for both frameworks, letting learners choose their preferred environment or compare implementations directly.

### Can I skip ahead to specific topics like computer vision or NLP?

Yes. While the learning path for AI For Beginners is organized linearly, the repository is intentionally modular. You can jump to [`lessons/4-ComputerVision/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/4-ComputerVision/README.md) or [`lessons/5-NLP/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/5-NLP/README.md) as long as you understand the foundational neural network concepts covered in [`lessons/3-NeuralNetworks/README.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/3-NeuralNetworks/README.md).

### Where is the course setup guide located?

The setup instructions are in [`lessons/0-course-setup/setup.md`](https://github.com/microsoft/AI-For-Beginners/blob/main/lessons/0-course-setup/setup.md). This file guides you through creating a conda environment, configuring VS Code, and optionally using a dev-container so that all subsequent notebooks run correctly.