Pinned Loading

1. jedi_noreward_rl jedi_noreward_rl Public

   Self-Learning Jedi AI: An AI that learns to be a Jedi on its own. All feedback comes from within the agent itself. Specificaly, it learns to predict the future and rewards itself for finding (image…

   Python 2

2. sign_nanoGPT sign_nanoGPT Public

   Forked from karpathy/nanoGPT

   nanoGPT with sign gradient descent instead of adamw

   Python 2

3. efficient_zero efficient_zero Public

   Forked from boularbahsmail/Asteroids-Game

   A quick implementation of MuZero (https://arxiv.org/abs/1911.08265) and EfficientZero (https://arxiv.org/abs/2111.00210) for the classic arcade game Asteroids.

   Python 1

4. elo elo Public

   Elo-based ranking system for Jedi Academy.

   C++

5. arbitrage arbitrage Public

   Single cryptocurrency exchange arbitrage via Bellman-Ford algorithm.

   Python

6. dqn dqn Public

   Deep Q-Network (deep reinforcement learning) applied to Tetris.

   Python