I learn new and fun stuffs in my spare time. For me, implementing algorithms is the best way to learn new concepts. Here, you can find some of them.
ReinforcementLearning.jl
I created this framework as a playground while I was studying the Reinforcement Learning. You can define Markov Decision Processes (MDPs) and custom environments. The framework also contains implementation of some of the algorithms (policy and value iteration, Q-Learning, Sarsa and Monte-Carlo search) to solve a given mdp.
Knet.jl
I am one of the contributors of Knet which is the KoƧ University deep learning framework implemented in Julia. It supports automatic differentiation using dynamic computational graphs for models defined in plain Julia.
ParticleSwarmOptimization.jl
I created this framework as a result of my project for the Parallel Programming course. Particle Swarm Optimization (PSO) is a nature inspired heuristic optimization method. This project contains an implementation of Particle Swarm Optimization in Julia. The repo provides serial and parallel implementation with MPI. The project also provides an interface for training neural networks using PSO with the help of Knet.
Gym.jl
A Julia wrapper of OpenAi gym This package allows you to use environments from gym within the Julia.
nature-inspired
Implementation of some nature-inspired algorithms including Genetic Algorithms, Ant Colony Optimization, and Particle Swarm Optimization.