semi-ergodic's repositories
RLlib-with-Dict-State
A minimal example demonstrating how to use RLlib with states which are presented as dictionaries
narrow-corridor-ai
A reinforcement learning project for crowd-dynamics in a very narrow corridor
simplest-world-Actor-Critic
Reinforcement learning, Policy Gradient, Actor-Critic, AC, Agent-based Simulation, Simple-world
recruiter-problem
The secretary problem is a problem that demonstrates a scenario involving optimal stopping theory.
cart-pole-deep-RL-actor-critic
Solving the inverted pendulum problem with deep-RL actor-critic (with shared network between the value-evaluation and the policy, epsilon-greedy policy). Some implementation issues concerning the stability are discussed.
epydemy-ai
Using the predictions of the agent-based simulation code epydemy, we train a deep neural-network to help identifying the individual susceptibile to catching the virus (the high-risk group). This deep neural-network enables us to find the quarantine-policy most effectively.
moo_as_soo
addressing multi-objective optimization as a single objective optimization with RL
secretary-problem-env
An environment compatible with open-AI gym for the secretary problem
simplest-world-REINFORCE
Reinforcement learning, Policy Gradient, REINFORCE, Agent-based Simulation, Simple-world
cart-pole-deep-RL-DDQN
Solving the inverted pendulum problem with deep-RL double DQN. Some implementation issues and tests are discussed.
communicative-MARL-v1
A multi-agent RL where the agents learn "what" to communicate with each other.
Furnace-Env
a furnace environment compatible with Gymnasium
multi-agent-trains-env
An environment (in openAI gym sense of the word) with multiple agents as a test bed for MA-RL algorithms
munching-recipes
web scraping www.allrecipes.com and returning info in form of Python dictionaries
acme
A library of reinforcement learning components and agents
anisotropic-very-very-simple-MD
[playground codes] This is a prototype MD (a one filer!!) code with anisotropic particle interactions
harmonic-oscillator-pinn
Code accompanying my blog post: So, what is a physics-informed neural network?
MagLorGasCpp
A code for magnetic Lorentz gas with obstacles which are made of polygon.
ray
Ray is a unified framework for scaling AI and Python applications. Ray consists of a core distributed runtime and a toolkit of libraries (Ray AIR) for accelerating ML workloads.
uneven_maze
a simple maze with an uneven surface