How Wayfair balances exploration and exploitation in AI
As an e-commerce owner, when deciding how to use the data you are collecting on your website it is common to ask yourself: do I exploit the information gathered from a previous user or do I explore information of the next user that comes to my site?
According to Tom Croonenborghs, Associate Director of Search & Recommendations at Wayfair, the answer is that it depends on the use case. This podcast will explain the difference between exploration and exploitation when it comes to a new user to help define how you can move forward.
In this podcast, Tom discusses how with an exploratory phase you can show to new users that you have not met, the best sellers and a wide overview of the catalog. Remember that in this phase you don’t know which product will peak the interest of that specific user.
Now with the exploitation phase, you have a clearer understanding of your user and their behavior with your products. As you might know your user a bit more than in the exploratory phase, you can get a better idea of their preferences to finally predict the product recommendations.
In summary, to find the right balance between exploitation and exploration using machine learning can be a whole task by itself. The most important point is that you feel confident on the data you own about your users and your products. Then you will be able to be implicit on your product recommendations to create a memorable user experience.
About Tom Croonenborghs
Tom Croonenborghs is the Associate Director of Search & Recommendations at Wayfair. He graduated with the degree of Master of Engineering in Computer science and another Master in Industrial Engineering, both in Belgium. The he pursued a Ph.D. on relational reinforcement learning. For already four years, he is settled in Boston at Wayfair, an American e-commerce company specializing in homewares and furniture. Together with his team, they focus on developing all kinds of recommendation strategies, such as automatically selecting the content we send to each of our customers in our marketing emails. He is responsible for developing roadmaps together with their stakeholders (business stakeholders, engineers, analysts, ...) and monitoring and steering the progress made by the different people in my team for the different problems.
Wayfair is the destination for all things home and a place that helps everyone, anywhere, create their feeling of home. One of the largest online home retailers in the world, a single site where people could find millions of products for every part of their homes. They are driven by a world-class team with the future always in mind – constantly striving to help everyone, anywhere, create spaces that are just right for them.
H1 - Heading 1
H2 - Heading 2
H3 - Heading 3
H4 - Heading 4
H5 - Heading 5
H6 - Heading 6
Paragraph - The rich text element allows you to create and format headings, paragraphs, blockquotes, images, and video all in one place instead of having to add and format them individually. Just double-click and easily create content.
A rich text element can be used with static or dynamic content. For static content, just drop it into any page and begin editing. For dynamic content, add a rich text field to any collection and then connect a rich text element to that field in the settings panel. Voila!
Bold - Headings, paragraphs, blockquotes, figures, images, and figure captions can all be styled after a class is added to the rich text element using the "When inside of" nested selector system.
Italic - Headings, paragraphs, blockquotes, figures, images, and figure captions can all be styled after a class is added to the rich text element using the "When inside of" nested selector system.