# One Embedding To Do Them All

arXiv: https://arxiv.org/abs/1906.12120 · HTML: https://mlsanity.com/paper/1906.12120/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Jun 2019.

## Built on by (most cited first, 10 of 10)

- [Unsupervised Extractive Summarization of Emotion Triggers](https://mlsanity.com/paper/2306.01444/) (2023)
- [Classification for everyone : Building geography agnostic models for fairer recognition](https://mlsanity.com/paper/2312.02957/) (2023)
- [Shaping Political Discourse using multi-source News Summarization](https://mlsanity.com/paper/2312.11703/) (2023)
- [Exploring Graph Based Approaches for Author Name Disambiguation](https://mlsanity.com/paper/2312.08388/) (2023)
- [Multimodal Temporal Fusion Transformers Are Good Product Demand Forecasters](https://mlsanity.com/paper/2307.02578/) (2023)
- [Sarcasm Detection in a Disaster Context](https://mlsanity.com/paper/2308.08156/) (2023)
- [Zero-shot Active Learning Using Self Supervised Learning](https://mlsanity.com/paper/2401.01690/) (2024)
- [Aligning Hotel Embeddings using Domain Adaptation for Next-Item Recommendation](https://mlsanity.com/paper/2108.13824/) (2021)
- [Detecting anxiety from short clips of free-form speech](https://mlsanity.com/paper/2312.15272/) (2023)
- [Exploring intra-task relations to improve meta-learning algorithms](https://mlsanity.com/paper/2312.16612/) (2023)

## Abstract (arXiv)

Online shopping caters to the needs of millions of users daily. Search, recommendations, personalization have become essential building blocks for serving customer needs. Efficacy of such systems is dependent on a thorough understanding of products and their representation. Multiple information sources and data types provide a complete picture of the product on the platform. While each of these tasks shares some common characteristics, typically product embeddings are trained and used in isolation. In this paper, we propose a framework to combine multiple data sources and learn unified embeddings for products on our e-commerce platform. Our product embeddings are built from three types of data sources - catalog text data, a user's clickstream session data and product images. We use various techniques like denoising auto-encoders for text, Bayesian personalized ranking (BPR) for clickstream data, Siamese neural network architecture for image data and combined ensemble over the above methods for unified embeddings. Further, we compare and analyze the performance of these embeddings across three unrelated real-world e-commerce tasks specifically checking product attribute coverage, finding similar products and predicting returns. We show that unified product embeddings perform uniformly well across all these tasks.