This recent article ["An Optimal Algorithm for Stochastic and Adversarial Bandits", Julian Zimmert, Yevgeny Seldin, 2018, arXiv:1807.07623] is really interesting. They quote our work on doubling trick, and propose what I believe is the first algorithm to be optimal for both settings. So impressive!

• I should read it carefully,
• And implement in SMPyBandits their algorithms,
• To do my own comparison against the state of arts algorithms,
• And check and verify their claims. (or disprove them?).