🏆 Threes! AI: deck-aware expectimax, N-tuple TD learning, and AlphaZero

TD-Gammon implementation

[ICLR 2020, Oral] Harnessing Structures for Value-Based Planning and Reinforcement Learning

Multi-Shot Approximation of Discounted Cost MDPs

Code for the Macro III course at the M.Sc/Ph.D programs at EPGE/FGV

RL with OpenAI Gym

GAN zoo include GAN, ACGAN, EBGAN, BEGAN, LSGAN, SAGAN, CVAE.

Flappy Bird for artificial intelligence/machine learning (Agent available: Q-Learning, SARSA, and combined with Backpropagation)

This is code from the Numerical Methods course at EPGE/FGV in 2018

EMINN implementation of Krusell-Smith model with neural network solutions.

optimal individualized treatment regime estimation for survival data with competing risks

The cognitive core required for superintelligence to emerge. All the ingredients assembled together here.

Goal Selection Strategies for Learning Goal-Oriented Value Functions