Learning Robust Models for e-Commerce Product Search
This addresses the issue of counterfactual biases in ranking algorithms for e-commerce search, improving customer experience, though it appears incremental as it builds on existing methods for handling noisy behavioral signals.
The paper tackles the problem of mismatched items in e-commerce product search, which degrades customer experience, by developing a deep end-to-end model that classifies mismatches and generates hard examples to improve robustness, achieving over 26% relative gain in F-score and over 17% in Area Under PR curve compared to baselines.
Showing items that do not match search query intent degrades customer experience in e-commerce. These mismatches result from counterfactual biases of the ranking algorithms toward noisy behavioral signals such as clicks and purchases in the search logs. Mitigating the problem requires a large labeled dataset, which is expensive and time-consuming to obtain. In this paper, we develop a deep, end-to-end model that learns to effectively classify mismatches and to generate hard mismatched examples to improve the classifier. We train the model end-to-end by introducing a latent variable into the cross-entropy loss that alternates between using the real and generated samples. This not only makes the classifier more robust but also boosts the overall ranking performance. Our model achieves a relative gain compared to baselines by over 26% in F-score, and over 17% in Area Under PR curve. On live search traffic, our model gains significant improvement in multiple countries.