Project Description: ai-by-hand-excel

What is the project about?

The project, "AI by Hand," provides a collection of Excel spreadsheets that implement various Artificial Intelligence (AI) and Deep Learning (DL) concepts and algorithms. It's designed to help users understand these concepts by manually working through the calculations.

What problem does it solve?

It demystifies complex AI/DL algorithms by breaking them down into step-by-step calculations within a familiar tool (Excel). This helps users grasp the underlying mechanics without getting bogged down in code. It bridges the gap between theory and practical implementation.

What are the features of the project?

• Basic Concepts: Softmax, LeakyReLU, Temperature.
• Advanced Networks/Techniques: Multi-Layer Perceptron (MLP), Backpropagation, Recurrent Neural Networks (RNNs), LSTMs (including Seq2Seq), xLSTMs, ResNet, Transformer (Simple and Full Stack), Self-Attention, Multihead Attention, Autoencoders, Mamba, AlphaFold.
• Fundamental Operations: Dot Product, Matrix Multiplication, Linear Layer.
• Lectures: DeepSeek (Multi-head Latent Attention + Mixture of Experts).
• Planned Additions: GAN, VAE, U-Net, CLIP, and more.

What are the technologies used in the project?

• Microsoft Excel (Spreadsheets).

What are the benefits of the project?

• Hands-on Learning: Provides a practical, interactive way to learn AI/DL concepts.
• Conceptual Clarity: Simplifies complex algorithms into manageable steps.
• No Coding Required: Accessible to users without programming experience.
• Familiar Environment: Uses Excel, a widely known and used tool.
• Visual Understanding: Spreadsheets offer a visual representation of the calculations.

What are the use cases of the project?

• Education: For students and learners to understand AI/DL fundamentals.
• Self-Study: For individuals interested in exploring AI/DL without coding.
• Teaching Aid: For educators to demonstrate AI/DL concepts in a visual manner.
• Prototyping: (Potentially) for quickly sketching out simple AI/DL models before implementing them in code.
• Debugging: To understand the step-by-step process of an algorithm for debugging purposes.