Michael Cukier's starred repositories
AutoSpotting
Saves up to 90% of AWS EC2 costs by automating the use of spot instances on existing AutoScaling groups. Installs in minutes using CloudFormation or Terraform. Convenient to deploy at scale using StackSets. Uses tagging to avoid launch configuration changes. Automated spot termination handling. Reliable fallback to on-demand instances.
LeducPoker
Python implementation of CFR In Leduc Hold'em
One-Card-Poker-CFG
Counterfactual Regret Minimization to find GTO solution to One Card Poker
Two-Card-Poker-CFR-AI
Using counter factual regret minimization to computer optimal ranges of hands for each decision
poker-agent-kit
Collection of game-theoretic algorithms for Poker
open_spiel
OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games.
NNDL-solutions
Solutions of the exercises and problems from Michael Nielsen's book Neural Networks and Deep Learning: http://neuralnetworksanddeeplearning.com/
math-as-code
a cheat-sheet for mathematical notation in code form
pluribus-poker-bot-js
An attempt to build the pluribus poker bot
Poker-Fictitious-Play
Fictitious Self-play & Reinforcement Learning
purge-wrangler
AMD & NVIDIA eGPUs for all Thunderbolt Macs.
COUNTERFACTUAL-REGRET-MINIMIZATION
Implemented the CFR+ and PureCFR algorithms in Python to find the optimal strategies to 2-player extensive-form games, which was also used in Libratus, the best poker AI in the world
deeper-stacker
DeeperStacker: DeepHoldem Evil Brother
regret-matching
Simple implementation of regret matching algorithm for RPS nash equilibrium computation via self-play
coms4995-finalproj
coms4995 Final Project Poker AI