SVM, FCN, Auto-encoder + MLP, GCN and EV-GCN

Address the JSP problem through DRL, including mlp, gcn, transformer policies.

A tutorial of Graph Neural Network, including PPT and ipynb file. We compared GNN with machine learning method SVM and classify neural network MLP. And analyse the different message propagate methods (like GCN, mean and sum) and the impact of the removal ratio of edges.

MLP vs GCN — a hands-on demo of why edges matter.

Sign Language Translator using MLP, CNN and GCN.

A systematic study comparing GCN and MLP performance across Homophilic (Cora) and Heterophilic (Texas, Chameleon) graph datasets. This repository serves as a foundational benchmark, illustrating the performance degradation of traditional GNNs in low-homophily environments, and synthesizes key research directions in Heterophilic Graph Learning.

perform three models (MLP, GCN, and GraphSAGE)  using the Cora dataset for node prediction.

This notebook explores node classification in a simulated celestial graph using PyTorch Geometric. It compares three models — MLP (baseline), GCN (message passing), and GAT (attention-based). Accuracy and embeddings are visualized to understand model differences.

A general autonomous ML research framework that designs and runs experiments, learns from history, and improves over time with minimal human intervention. Supports classification, regression, recommendation, ranking, and graph learning across model families (MLP, PMF, NeuralMF, GCN, GraphSAGE, GAT).

This project presents a comparative analysis of MLP, GCN, GAT, and an advanced GAT model on the UNSW-NB15 cybersecurity dataset. It applies graph-based learning techniques for detecting anomalous network activity using node-level classification and feature engineering.

MLP&GCN

No description available

No description available

Bachelor assignment

No description available

No description available

No description available

GNN-based credit card fraud detection (PyTorch Geometric). k-NN graph + GCN vs MLP baseline.

A lightweight library for Multi-view learning experiments, featuring standardized preprocessing and several baseline implementations including MLP, GCN, and Transformer.

Hierarchical MARL (H-MARL) for Warehouse Automation: Scalable pathfinding and task allocation using GCN Managers, MLP Workers, and CBF Safety Layers in RWARE.

A robust, real-time Yoga Pose Classification system using Graph Convolutional Networks (GCN) with residual connections, LSTM for temporal modeling, and MLP for classification. Built with MediaPipe, PyTorch Geometric, and Streamlit, the system provides accurate pose prediction along with real-time correction suggestions.

Comprehensive benchmarking of group-fair graph neural networks across five real-world datasets (NBA, Income, Recidivism, Pokec_z, Pokec_n). This project integrates and evaluate models: FairGNN, BIND, NIFTY, MLP, and Vanilla GCN to investigate algorithmic approaches to mitigating group bias in graph-based learning systems

Comprehensive benchmarking of group-fair graph neural networks across five real-world datasets (NBA, Income, Recidivism, Pokec_z, Pokec_n). This project integrates and evaluate models: FairGNN, BIND, NIFTY, MLP, and Vanilla GCN to investigate algorithmic approaches to mitigating group bias in graph-based learning systems