Cold Lazarus is my on-going passion project being created in the Unity Game Engine. I program, draw, animate, compose and everything else which goes into making a video game from scratch!
About The Game: Cold Lazarus is a 2D adventure platformer with the charm of Banjo-Kazooie and mechanics of a Super Metroid. Follow our protagonist, Lazarus the Do Do bird, as he unravels the mysteries and turmoil of a strange new world. Chaos echoes across the land and cuddly characters are the innocent casualties; as the great powers fight over new found technologies left behind by the ancients. The power struggle is real, the technology is dangerous! Scary rifts to spooky dimensions are opening up! Monsters are running havoc! Hopefully our amnesia stricken friend Lazarus can help.
Project Summary: “Nobody is Attacking Me!” is a real-time tower defense virtual-reality game, where the player is the tower! The player takes on the form of a giant cyclops, towering over mythological landscapes. As beautiful as it is in this game, fighting off waves of frightened Greek soldiers is your prerogative. Perhaps you’re just misunderstood, but that’s not stopping the hordes of Greek armies from launching rocks at you from their deadly catapults. Don’t get careless! Their throwing spears are very sharp! You don’t want to lose your one and only eye! Retaliate in fury by picking up the soldiers slashing at your legs and hurling them like bowling balls towards their comrades!
Amazeing Meatball is a fully developed cell phone game for the Android OS and can be downloaded from the Google Play store!
Meatball the Dog is lost and needs to find his way home! Help Meatball navigate over One Hundred levels of Increasing difficulty. Each map is randomly generated providing unique experiences every time. Avoid dangerous enemies, complete various challenges and find the the exit! Each level completed brings Meatball one step closer to home.
Read about the adventure as Meatball chronicles his journey in his best friend “Diary”. Meatballs emotions, memories and turmoil’s all laid out for an epic tale. Join Meatball’s journey in the “aMAZEing Meatball.” This U-net style deep learning model was trained to categorize individual pixels of an image for classification in the highway driving dataset.
Using Unity’s ML-Agents toolkits, I was able to simulate bird swarm behavior using machine learning. Through curriculum learning in an adaptive environment, ‘Bird Swarm Agents’ with a set of actions and rewarded behavior, learned to quickly navigate towards a target while avoiding collisions with each other. The results were surprising with behavior matching closely with bird swarms in nature.
Ticket to Ride is a board game where 2 – 5 players compete to build rail lines across the United States. Each player is dealt destination tickets with the goal of connecting two cities across the board.
Monte Carlo Tree Search (MCTS) utilizes Monte Carlo Simulation in order to generate node values/estimates to help guide a state based search. MCTS focuses on paths which appear to be more promising, narrowing the amount of nodes needed to visit in order to find a ‘winning solution.’
Employing a semantic segmentation approach and building off of the Xception Style U-Net, this deep learning neural net categorizes individual pixels for image classification on the highway driving dataset.
The intended purpose of this research was to see if a classification model could be implemented to help predict cryptocurrency prices. The simple classification approach aimed to predict 3 classes; Increase or Decrease above or below a specified threshold; or no change. The goal was to predict (n) minutes into the future using the previous (n) minutes of recorded data. This was a semester project for my Deep Learning Course.
I built a Connect Four AI which utilized a min-max search tree with alpha-beta pruning, with a modified monte-carlo simulation for heuristics.
“Deeds will not be less valiant because they are unpraised.”
CyrusTheVirus Studios 