This project explores an intelligent approach to clustering using generative AI. By leveraging a convolutional autoencoder, the framework learns meaningful latent representations of raw signals, capturing subtle patterns that traditional feature extraction methods may overlook. An adaptive clustering pipeline then optimizes cluster assignments by balancing reconstruction accuracy, cluster compactness, and separation. The result is a powerful framework for automatically discovering and analyzing distinct input behaviors, transforming complex data into interpretable insights. More to come!

Alireza Rafiei
AI Enthusiast & Lifelong Learner!
- Atlanta, GA, USA
- Emory University
- Google Scholar
- ResearchGate
- Github