Saved in:
Bibliographic Details
Main Authors: Yang, Wenchuan, Yang, Cheng, Li, Jichao, Tan, Yuejin, Lu, Xin, Shi, Chuan
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.06925
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917823670910976
author Yang, Wenchuan
Yang, Cheng
Li, Jichao
Tan, Yuejin
Lu, Xin
Shi, Chuan
author_facet Yang, Wenchuan
Yang, Cheng
Li, Jichao
Tan, Yuejin
Lu, Xin
Shi, Chuan
contents The personalized bundle generation problem, which aims to create a preferred bundle for user from numerous candidate items, receives increasing attention in recommendation. However, existing works ignore the order-invariant nature of the bundle and adopt sequential modeling methods as the solution, which might introduce inductive bias and cause a large latency in prediction. To address this problem, we propose to perform the bundle generation via non-autoregressive mechanism and design a novel encoder-decoder framework named BundleNAT, which can effectively output the targeted bundle in one-shot without relying on any inherent order. In detail, instead of learning sequential dependency, we propose to adopt pre-training techniques and graph neural network to fully embed user-based preference and item-based compatibility information, and use a self-attention based encoder to further extract global dependency pattern. We then design a permutation-equivariant decoding architecture that is able to directly output the desired bundle in a one-shot manner. Experiments on three real-world datasets from Youshu and Netease show the proposed BundleNAT significantly outperforms the current state-of-the-art methods in average by up to 35.92%, 10.97% and 23.67% absolute improvements in Precision, Precision+, and Recall, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2406_06925
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Non-autoregressive Personalized Bundle Generation
Yang, Wenchuan
Yang, Cheng
Li, Jichao
Tan, Yuejin
Lu, Xin
Shi, Chuan
Machine Learning
Information Retrieval
The personalized bundle generation problem, which aims to create a preferred bundle for user from numerous candidate items, receives increasing attention in recommendation. However, existing works ignore the order-invariant nature of the bundle and adopt sequential modeling methods as the solution, which might introduce inductive bias and cause a large latency in prediction. To address this problem, we propose to perform the bundle generation via non-autoregressive mechanism and design a novel encoder-decoder framework named BundleNAT, which can effectively output the targeted bundle in one-shot without relying on any inherent order. In detail, instead of learning sequential dependency, we propose to adopt pre-training techniques and graph neural network to fully embed user-based preference and item-based compatibility information, and use a self-attention based encoder to further extract global dependency pattern. We then design a permutation-equivariant decoding architecture that is able to directly output the desired bundle in a one-shot manner. Experiments on three real-world datasets from Youshu and Netease show the proposed BundleNAT significantly outperforms the current state-of-the-art methods in average by up to 35.92%, 10.97% and 23.67% absolute improvements in Precision, Precision+, and Recall, respectively.
title Non-autoregressive Personalized Bundle Generation
topic Machine Learning
Information Retrieval
url https://arxiv.org/abs/2406.06925