At this hackathon, local communities of developers and researchers united to tackle some fundamental and daunting challenges of quantum coding within 24 hours. This invite-only event hosted 40 people from Madrid and the nearby Basque Country. Backgrounds of the attendees ranged from computer science to physics and IT to graphic design. Overall, four universities, 10 developer communities, four companies and members of the Spanish National Research Council were represented.
After heated deliberation, the judges determined three winning projects. Quantum Reinforced Learning (QRL) won for scientific excellence and Q-means for community excellence. The overall winner, excelling in the category of innovation, was Qonway’s Game of Life.
Pablo Moreno, Santiago Varona together with Ana Martín, Alfredo Ibias and Pablo Barrio, worked on the Hackathon project consisted of experimentally quantifying one of the most basic algorithms of Reinforcement Learning (Reinforcement Learning) within the field of what is known as Machine Learning. The algorithm in question is known as Policy Iteration, and consists of making successive iterations in which the policy of the agent that is learning, that is, the action taken in each possible state, is improving to achieve the objective. The quantization comes from the fact that we have to solve a system of linear equations, for which the quantum algorithm HHL is a very good option. The experiment in particular consisted of applying the algorithm to an environment in the form of a chessboard known as gridworld, in which the agent (a square) must move to reach its goal.