Safe multi-armed bandit implementations:

• Eps-Greedy (fixed rate, inverse squared rate)
• UCB family (UCB1, α-UCB, (α, ϕ)-UCB) [1]
• Exp3 (hyperparameter-free rate from [1])
• Exp4.R [2]

  nix-build /path/to/hbandit/or/url/to/tarball -A hbandit.doc

[1] Bubeck, S. et al. 2012. Regret analysis of stochastic and nonstochastic multi-armed bandit problems. Foundations and Trends in Machine Learning. 5, 1 (2012), 1–122.

[2] Sun, W. et al. 2017. Safety-aware algorithms for adversarial contextual bandit. Proceedings of the 34th international conference on machine learning-volume 70 (2017), 3280–3288.