[1]
M.- hwan Oh and G. Iyengar, “Multinomial Logit Contextual Bandits: Provable Optimality and Practicality”, AAAI, vol. 35, no. 10, pp. 9205-9213, May 2021.