https://github.com/ilke-kas/probabilistic-graphic-models
A portfolio of advanced projects in probabilistic graphical models (PGMs), featuring hands-on notebooks for model design, inference, and real-world applications. Includes work on Bayesian networks, image denoising, and creative PGM explorations. Ideal for data science, machine learning, and AI.
https://github.com/ilke-kas/probabilistic-graphic-models
ai bayesian-networks data-science inference machine-learning probabilistic-graphical-models python
Last synced: about 1 month ago
JSON representation
A portfolio of advanced projects in probabilistic graphical models (PGMs), featuring hands-on notebooks for model design, inference, and real-world applications. Includes work on Bayesian networks, image denoising, and creative PGM explorations. Ideal for data science, machine learning, and AI.
- Host: GitHub
- URL: https://github.com/ilke-kas/probabilistic-graphic-models
- Owner: ilke-kas
- Created: 2025-10-05T05:05:19.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-10-06T07:26:02.000Z (8 months ago)
- Last Synced: 2025-10-06T09:37:09.562Z (8 months ago)
- Topics: ai, bayesian-networks, data-science, inference, machine-learning, probabilistic-graphical-models, python
- Language: Jupyter Notebook
- Homepage:
- Size: 8.33 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Probabilistic Graphical Models — Project Portfolio
Welcome! This repository showcases a collection of advanced, hands-on projects exploring the theory and application of probabilistic graphical models (PGMs). Each notebook is designed to be accessible to both technical and non-technical audiences, with clear explanations, visualizations, and practical code. These projects demonstrate expertise in probabilistic modeling, inference, and real-world problem solving — skills highly valued in data science, machine learning, and AI-driven industries.
---
## Project Notebooks
### 1. Probabilistic Graphical Models — Exploratory Notebook
**File:** `probabilistic-graphical-models-exploratory.ipynb`
A comprehensive, research-style notebook introducing the foundations of probabilistic graphical models. This notebook covers:
- Conditional independence and d-separation
- Model complexity and parameterization
- Bayesian networks and factor graphs
- Inference techniques and practical examples
- Visual explanations and mathematical derivations
Ideal for readers seeking a deep yet accessible introduction to PGMs, with a blend of theory and hands-on code.
---
### 2. Probabilistic Graphical Models — Image Denoising Project
**File:** `probabilistic-graphical-models-image-denoising.ipynb`
A practical project applying PGMs to the challenging task of image denoising. Highlights include:
- Formulating image denoising as an energy minimization problem
- Implementing factor graphs and inference algorithms
- Experimenting with real and synthetic noisy images
- Visualizing the denoising process and energy landscape
This notebook demonstrates how probabilistic modeling can solve real-world computer vision problems, with clear code and results.
---
### 3. Probabilistic Graphical Models — Inference Project
**File:** `probabilistic-graphical-models-inference-project.ipynb`
A deep dive into inference algorithms and structure learning in PGMs. Key topics:
- Exact and approximate inference methods
- Structure learning from data
- Applications to real datasets
- Step-by-step code and visualizations
This project is perfect for those interested in the mechanics of inference and learning in graphical models, with a focus on practical implementation.
---
### 4. PGM Explorations
**File:** `pgm-explorations..ipynb`
A curated set of advanced explorations and experiments in probabilistic graphical models. This notebook features:
- Creative applications and novel model structures
- Comparative studies of inference techniques
- Insights into the strengths and limitations of PGMs
Great for readers looking to see PGMs in action beyond standard textbook examples, with a focus on innovation and critical thinking.
---
## Why Probabilistic Graphical Models?
Probabilistic graphical models are a cornerstone of modern AI, enabling interpretable, flexible, and powerful reasoning under uncertainty. The skills demonstrated in these projects — from model design to inference and application — are directly relevant to roles in data science, machine learning engineering, and research.
If you have questions or would like to discuss these projects, feel free to reach out!