Artwork Personalization at Netflix
Netflix Technology Blog

very impressive and helpful article! Thanks~

I was wondering what contextual bandit algorithm did you use? Is it LinUCB, or other UCB type contextual bandit algorithms? As I know, many of them can not really be updated online as the computation expense is tooo high(e.g. the inverse of a big matrix in each update). Would appreciate it i you could provide some hints. Thanks again.

Like what you read? Give Jenkins yu a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.