# Discrete Mathematics in AI: Function Impact & Mapping
> Explore how one-one, onto, and bijective functions control AI decisions, identity preservation, and data compression in machine learning models.

Tags: discrete-mathematics, ai-logic, machine-learning, function-theory, data-science, biometrics, mathematics
## Calculus vs. Discrete Math in AI
- Contrast between continuous gradients and discrete set decisions in AI.
- Introduction to one-one, onto, and bijective mapping.

## Why Discrete Math Controls AI
- AI classifiers operate as discrete functions: f: A → B where A and B are finite sets.
- Decisions involve discrete labels, IDs, tokens, and classes.

## One-One (Injective) Functions
- **Definition:** Each input maps to a unique output; no two elements share the same image.
- **AI Impact:** Preserves identity. Failure (non-injectivity) leads to 'Discrete Collisions' in systems like face recognition.

## Onto (Surjective) Functions
- **Definition:** Every element in the output set (B) has at least one source.
- **AI Impact:** Ensures coverage of reality. Failure leads to 'model blindness' where certain labels (e.g., rare diseases) are never predicted.

## Bijective Functions: Perfect Pairing
- **Definition:** Both one-one and onto. Reversible mapping (Encryption).
- **The Autoencoder Problem:** Most AI models are not bijective; they compress data, leading to noise and hallucinations.

## Discrete Comparison Matrix
- **One-One:** No collisions, allows unused outputs, not reversible. Use case: IDs.
- **Onto:** Collisions allowed, no unused outputs, not reversible. Use case: Classifiers.
- **Bijective:** No collisions, no unused outputs, reversible. Use case: Encryption.
---
This presentation was created with [Bobr AI](https://bobr.ai) — an AI presentation generator.