Training a Deep Neural Network

Project: Handwritten Digit Recognition with Deep Neural Networks (CS467 - Intro to Reverse Engineering)
In this project, I built a deep neural network from the ground up using PyTorch to recognize handwritten digits from the MNIST dataset. I implemented a custom classifier with multiple hidden layers, optimized it using backpropagation, and tested its accuracy using real image data. When early predictions failed, I debugged model weights, explored data distribution, and adjusted layer configurations—showcasing my problem-solving mindset and persistence in figuring out why things weren’t working and how to fix them. This hands-on project reflects my ability to break down complex systems, analyze behavior, and apply logical thinking to reverse engineer solutions effectively.