Page Archive

Page Not Found

Dr. Ozgur Ural

Ph.D. in Machine Learning (ERAU 2025). Four IEEE Access papers on trustworthy ML, proof-of-learning, and model watermarking. Senior software engineer.

Advanced Task and Packet Management in CPP

Published: December 09, 2022

Adversarial examples for Proof of Learning

Published: March 02, 2024

Archive Layout with Content

Awesome leading and managing

Published: December 22, 2022

Technical Writing

Posts by Category

ChampionshipVisualizer

Published: July 23, 2023

Posts by Collection

CS 595A TEAM 13 DATA

Published: March 10, 2022

CS 595A TEAM 13

Published: March 10, 2022

Essays

Slow, indirect writing about machine learning, security, and the systems we are quietly building.

InterProcess and IntraProcess Messaging in CPP

Published: December 24, 2022

Gradient Pinball — Optimization, playable

Roll a ball down a high-dimensional loss landscape. Sweet spot or side valley — the math has opinions.

Proof-of-Learning (SecurePoL) — Playable Thought Experiment

Prove a model was trained, not downloaded. The loss-curve fingerprint is almost impossible to forge after the fact.

Block Race — Bitcoin consensus, playable

Mine, attack, or defend Bitcoin. The math is Nakamoto’s whitepaper §11; the bad decisions are entirely yours.

Redundancy Reactor — Triple-modular redundancy, playable

Triple-modular redundancy, correlation that bites, and the Ariane-5 story explained as a slider.

Model Heist Detector — AI watermarks, playable

Spread a faint statistical signature across thousands of weights and read it back through the noise.

Lab — five playable thought experiments

Five interactive thought experiments from my research: Bitcoin consensus, ML security, fault tolerance, optimization. Pick a real scenario, set your strategy, and let the math grade your aim.