Machine Learning Times
Machine Learning Times
EXCLUSIVE HIGHLIGHTS
Effective Machine Learning Needs Leadership — Not AI Hype
 Originally published in BigThink, Feb 12, 2024.  Excerpted from The...
Today’s AI Won’t Radically Transform Society, But It’s Already Reshaping Business
 Originally published in Fast Company, Jan 5, 2024. Eric...
A University Curriculum Supplement to Teach a Business Framework for ML Deployment
    In 2023, as a visiting analytics professor...
The AI Playbook: Providing Important Reminders to Data Professionals
 Originally published in DATAVERSITY. This article reviews the new...
SHARE THIS:

7 months ago
Leveraging Transformers to Improve Product Retrieval Results

 
Originally published in Amazon Science, Aug 3, 2023.

Assessing the absolute utility of query results, rather than just their relative utility, improves learning-to-rank models.

When a customer clicks on an item in a list of product-search results, it implies that that item is better than those not clicked. “Learning to rank” models leverage such implicit feedback to improve search results, comparing clicked and unclicked results in either “pairwise” (comparing pairs of results) or listwise (judging a results position within the list) fashion.

A problem with this approach is the lack of absolute feedback. For instance, if no items in the selection are clicked, it’s a signal that none of the results was useful. But without clicked items for comparison, learning-to-rank models can do nothing with that information. Similarly, if a customer clicks on all the items in a list, it could indicate that all the results were useful — but it could also indicate a fruitless search to find even one useful result. Again, learning-to-rank models can’t tell the difference.

In a paper we’re presenting at this year’s International Conference on Knowledge Discovery and Data Mining (KDD), we describe a new approach to learning to rank that factors in absolute feedback. It also uses the type of transformer models so popular in natural-language processing to attend to differences among items in the same list to predict their relative likelihood of being clicked.

To continue reading this article, click here.

5 thoughts on “Leveraging Transformers to Improve Product Retrieval Results

  1. In the bitlife Life Simulator, you will play a simulation game where you must make life-changing choices. For instance, you could wed the love of your life, start a family, and further your education. You can also choose things that, to be honest, will frighten your parents, but hey, maybe they’ll be entertaining.

     

Leave a Reply