"GNN (Graph Neural Network) ML Learning" (1) Curriculum - Approach

Shifting Perceptions and Expectations of AI from Companies

The way companies perceive and expect from AI is evolving.

Until now, aside from cutting-edge research and product development, the role of AI engineers and data scientists in enterprises can generally be categorized as follows:

1. Image Processing: Companies have widely utilized CNN-based models for object recognition and comparison (performance improvement). To overcome data scarcity, techniques like few-shot learning have been applied in areas such as product search, similar image/product recommendation, and design generation.

2. Text Processing: Enterprises have leveraged various Transformer-based models trained on proprietary datasets for customer intent recognition and sentiment analysis.

3. Data-Driven Decision Support: Companies often assume they possess sufficient data, but in many cases, they lack the necessary datasets. AI teams analyze available data, form hypotheses, and generate insightful reports to support and persuade decision-makers.

4. Data Collection & Labeling: One of the ongoing challenges is securing training datasets and labeled data for model development. AI teams are often tasked with curating and annotating data for continuous improvement.

However, as major corporations establish their own AI Centers (or AI Innovation Labs)—with many of these initiatives now entering their third year—decision-makers are beginning to realize that simply copying and pasting reference service models or implementing SOTA (State-of-the-Art) models from books and research papers is not enough to achieve their market objectives for products and services.

This shift presents both a challenge and an opportunity for AI professionals.
As companies seek differentiation beyond pre-built AI solutions, there will be a growing demand for AI specialists who can innovate, customize, and deploy tailored AI models that align with unique business needs.

<What is Driving the Rise in Popularity of GNNs?>

In my case, to stay ahead of these changes and avoid falling behind, I have decided to develop a product by applying Multi-Modal Networks and GNN in the fashion and education sectors, which I have recently started exploring

<1. First, build a GNN to predict fashion and education market trends.>

To achieve this, I plan to prepare a curriculum focused on acquiring the knowledge and technical skills necessary for AI service planning and the design and implementation of applicable models in the fashion and edtech markets.

<2. Learning Variants of GNN>

(I couldn't enroll in the relevant courses during my graduate program... 😢 Looks like it's going to be a tough journey.)

Main Learning Keywords

• Machine Learning with Graph Networks
• Graph Network Design
• Mathematical Operations
• Model Architecture & Implementation
• Using Reinforcement Learning to Predict Links (Relationships) in Generated Graph Networks

What Are the Essential Learning Resources, and What Approach Should I Take?

I will organize the key materials and the sequential approach for learning.

---

Learning Outline

1. Review of Graph Neural Networks: Methods and Applications
2. A Comprehensive Study on Graph Embedding: Challenges, Techniques, and Applications
3. Graph Embedding Techniques, Applications, and Performance
4. Network Embedding
5. Attention Models for Graphs
6. Deep Learning for Network Biology
7. Representation Learning on Graphs: Methods and Applications
8. Network Representation Learning
9. Graph Summarization Techniques and Applications
10. Must-Read Papers on Knowledge Representation Learning (KRL) / Knowledge Embedding (KE)
11. Node2Vec
12. Prediction Analysis with Neo4j and TensorFlow
13. Knowledge Graph Embedding: Approaches and Applications
14. A Novel Embedding Model for CNN-Based Knowledge Base Completion
15. GEMSEC: Graph Embedding with Self-Clustering
16. Relational Inductive Bias in Graph Networks
17. Convolutional Graph Networks
18. GraphSAGE
19. Smart Reply: Automated Response Suggestions for Emails
20. 3D Graph Neural Networks for RGBD Semantic Segmentation
21. DeepPath: A Reinforcement Learning Approach for Knowledge Graph Reasoning
22. Multi-Hop Knowledge Graph Reasoning with Reward Shaping
23. Neural Tensor Networks
24. MacGraph — Iterative Reasoning for Knowledge Graphs
25. KBGAN: Adversarial Learning for Knowledge Graph Embedding
26. Constructivist Networks for Machine Reasoning
27. Graph Classification with Structural Attention
28. GAMEnet: Graph-Augmented Memory Networks for Drug Combination Recommendation
29. Modeling Relational Data Using Graph Convolutional Networks
30. Answering Questions with Knowledge Graphs and Sequence Translation

---

🚀 Approach:
✅ Start with foundational knowledge (Graph Neural Networks & Embeddings)
✅ Progress to advanced techniques (Reinforcement Learning, Attention Models, and GANs in Graphs)
✅ Explore real-world applications (Healthcare, Smart Reply, Knowledge Graphs)
✅ Hands-on practice with tools like Neo4j, TensorFlow, and PyTorch Geometric

This structured learning path ensures a comprehensive understanding of Graph Neural Networks and their practical applications in AI-driven tasks.

Background Knowledge & Lectures

For a solid foundation in Graph Neural Networks (GNNs) and their applications, the following resources are recommended:

---

📚 References & Learning Materials

1️⃣ CS224W: Machine Learning with Graphs (Stanford University)

• A comprehensive course covering graph theory, graph embeddings, and deep learning with graphs.
• Topics include Graph Convolutional Networks (GCN), Graph Attention Networks (GAT), and applications in real-world problems.

2️⃣ How to Get Started with Machine Learning on Graphs

• A beginner-friendly guide introducing key concepts in graph-based machine learning.
• Explains fundamental techniques, including graph embeddings and relational learning.

3️⃣ Graph Neural Network Explanation (YouTube)

• A visual breakdown of GNN concepts, architectures, and practical implementations.
• Provides insights into how GNNs process non-Euclidean data structures.

4️⃣ Additional Lecture Materials

• Supplementary course slides and learning materials for deep diving into GNNs.