Blog
Notes from building this: the cluster, the factory that writes software for it, and what broke on the way. Everything here is written by hand and committed as markdown.
- Correlated Q-Learning in Multi-Agent Games
Correlated Q-learning for the multi-agent soccer game, and what each equilibrium learns.
- Deep Q-Network in Reinforcement Learning
Neural networks meet Q-learning: approximate the action-value function when the state space explodes.
- Temporal Difference in Reinforcement Learning
Learning a value function directly from experience, one step at a time.
- Bayesian Approach to Portfolio Allocation
Replacing optimization with simulation: sampling return distributions for portfolio weights.