Rafael Sterzinger's repositories
AlphaTrader
An implementation of a stock market trading bot, which uses Deep Q Learning
AlphaTrader-Web-Application
This is a web application for my AlphaTrader project which offers easy usage of my trained model
etmira-segmentation
Code for "Drawing the Line: Deep Segmentation for Extracting Art from Ancient Etruscan Mirrors" published @ ICDAR2024
ML-Exercise-1
A comparision of different machine learning techniques for regression
pytorch-maml-rl
Reinforcement Learning with Model-Agnostic Meta-Learning in Pytorch
QuadTree-Exercise
An implementation of the QuadTree datastructure to find trainstations and airports in a certain radius and also to find e.g. trainstations with X airports in a certain radius.
Strategy-Game-Programming
This project implements an agent which knows how to play mancala. It implements an Alpha-Beta-Pruning approach as well as a Monte Carlo Search Tree approach with a Deep Neural Network
Synthetic-Data-Generator
An implementation of a GAN which acts as a synthetic data generator for our ML-Excercise-3
Bachelor-Thesis
Bachelor's thesis on "Weighting Intrinsic and Extrinsic Rewards in Reinforcement Learning"
Huber-Verkehrsdaten
Android App for Public Transports in Vienna
OctoDash
OctoDash is a 2D co-op game where two players control an octopus, navigating sea-themed levels by steering its tentacles. Work together to reach your feast, using power-ups and interacting with the environment. Best played with a friend!
CodingDojo
This Coding Dojo gives a simple introduction to Neural Networks and was prepared for a summer internship 2019
etmira-interaction
Code for "Fusing Forces: Deep-Human-Guided Refinement of Segmentation Masks" published @ ICPR2024
macaw
Code for "Offline Meta-Reinforcement Learning with Advantage Weighting" [ICML 2021]
ML-Exercise-2
A comparision of different machine learning techniques for classification
PAI-Projects
Repository for Probabilistic AI, featuring implementations of Gaussian Process Regression, Bayesian Neural Networks, Bayesian Optimization, and Actor-Critic Reinforcement Learning.
SudokuSolver
A simple Sudoku solver written in Java