Articles

A player faces a grid board game with multiple coins. At random, a number K of coins will be selected. Find the expected number of steps the player will need to take to collect all coins. The player may start at any coin and can only move up, down, right or left. All coins are guaranteed to be reachable from all other coins.

The end tends to be much sexier than the beginning. Think about it, when we set ourselves a new challenge, we immediately think about the most gratifying aspect of it. These are some lessons from my story of how starting from the end when building my startup, Lagrange Systems, led us to failure.

Learn how to build an unbeatable TicTacToe robot! Implement a simple Reinforcement Learning agent that uses Exhaustive Search to find the best possible action.

A traceback of the beginnings of the collapse of Venezuela's economy, using national data that suggests that it began at some point between 1997 and 2007. For the full paper see: http://www.bcv.org.ve/publicaciones/ndeg-175-cambios-estructurales-de-los-sectores-productivos-de-la-economia-venezolana

An introductory reinforcement learning problem explained.

Deep Learning architectures are getting harder to design, but evolutionary algorithms may help us overcome this.