AI Case Study
Netflix improves core metrics by customising the product images displayed to users with machine learning
Netflix displays images from its video products in an online catalogue format for customers to choose from. It was hypothesised that personalising the image displayed to each unique customer based on their watching history would entice users to view videos more. Using contextual bandits, the company determined personalisation worked and could generate core metric lift.
Consumer Goods And Services
Entertainment And Sports
Netflix implements a variety of machine learning techniques to personalise its imagery for customers as well as then evaluating the results of personalisation against non-personalised image choices. "We ultimately ran an A/B test to compare the most promising personalized contextual bandits against unpersonalized bandits. This project is the first instance of personalizing not just what we recommend but also how we recommend to our members."
Netflix's hypothesis about the effects of personalisation was confirmed: "the personalization worked and generated a significant lift in our core metrics".
In-house; "For artwork personalization, the specific online learning framework we use is contextual bandits. Rather than waiting to collect a full batch of data, waiting to learn a model, and then waiting for an A/B test to conclude, contextual bandits rapidly figure out the optimal personalized artwork selection for a title for each member and context. Briefly, contextual bandits are a class of online learning algorithms that trade off the cost of gathering training data required for learning an unbiased model on an ongoing basis with the benefits of applying the learned model to each member context... In this online learning setting, we train our contextual bandit model to select the best artwork for each member based on their context. We typically have up to a few dozen candidate artwork images per title. To learn the selection model, we can consider a simplification of the problem by ranking images for a member independently across titles. ...These can be supervised learning models or contextual bandit counterparts with Thompson Sampling, LinUCB, or Bayesian methods that intelligently balance making the best prediction with data exploration....Once the model is trained as above, we use it to rank the images for each context. The model predicts the probability of play for a given image in a given a member context. We sort a candidate set of images by these probabilities and pick the one with the highest probability. That is the image we present to that particular member....To evaluate our contextual bandit algorithms prior to deploying them online on real members, we can use an offline technique known as replay".
Marketing Research Planning
"One challenge of image personalization is that we can only select a single piece of artwork to represent each title in each place we present it. In contrast, typical recommendation settings let us present multiple selections to a member where we can subsequently learn about their preferences from the item a member selects. ... What we seek to understand is when presenting a specific piece of artwork for a title influenced a member to play (or not to play) a title and when a member would have played a title (or not) regardless of which image we presented. Therefore artwork personalization sits on top of the traditional recommendation problem and the algorithms need to work in conjunction with each other... Another challenge is to understand the impact of changing artwork that we show a member for a title between sessions. Does changing artwork reduce recognizability of the title and make it difficult to visually locate the title again, for example if the member thought was interested before but had not yet watched it? Beyond the artwork for other titles, the effectiveness of the artwork for a title may depend on what other types of evidence and assets (e.g. synopses, trailers, etc.) we also display for that title. ...Finally, there are engineering challenges to personalize artwork at scale. One challenge is that our member experience is very visual and thus contains a lot of imagery. So using personalized selection for each asset means handling a peak of over 20 million requests per second with low latency. Such a system must be robust: failing to properly render the artwork in our UI brings a significantly degrades the experience."
User viewing history