Dylan Cope's Projects
Implementation of a rudimentary 3D software renderer.
TF-Agents is a library for Reinforcement Learning in TensorFlow
My experiments based of the paper "AI Safety via Debate"
A challenging Android arcade game.
Notes for the Bio-Inspired Artificial Intelligence course at the University of Bristol
Repository for computational neuroscience course
An interactive environment where protozoa-like entities live, die, and have offspring whose behaviours and morphologies can change.
An Android game where you swim through pipes under the ocean.
Playful visualisations of the Mandelbrot set
A GUI where users can input mathematical functions to be viewed and interacted with.
Applying the matrix equation solution to the least-squares problem in Java.
A simulation that models gravitational interactions between point particles.
A basic 2D maze environment where an agent start from the top left corner and try to find its way to the bottom left corner.
Porting Lattice-Boltzmann code to OpenCL
Small reference notebook for creating an image autoencoder with Keras
Notebook implementing a custom bimodal transformer architecture for masked language modelling in order to learn to caption COCO images.
Summary and Implementations of the content of the University of Bristol Image Processing and Computer Vision course.
Language Modelling with Transformers
A precise simulation of linear momentum between particles.
The aim of this project is to create a piece of software that parses a string as a mathematical function, and is able to perform various manipulations upon said function.