Everyone's preference changes differently: A weighted multi-interest model for retrieval

H Shi, Y Gu, Y Zhou, B Zhao, S Gao… - … on Machine Learning, 2023 - proceedings.mlr.press
User embeddings (vectorized representations of a user) are essential in recommendation
systems. Numerous approaches have been proposed to construct a representation for the …

Continuous-time convolutions model of event sequences

V Zhuzhel, V Grabar, G Boeva, A Zabolotnyi… - arXiv preprint arXiv …, 2023 - arxiv.org
Massive samples of event sequences data occur in various domains, including e-commerce,
healthcare, and finance. There are two main challenges regarding inference of such data …

Universal representations for financial transactional data: embracing local, global, and external contexts

A Bazarova, M Kovaleva, I Kuleshov… - arXiv preprint arXiv …, 2024 - arxiv.org
Effective processing of financial transactions is essential for banking data analysis.
However, in this domain, most methods focus on specialized solutions to stand-alone …

No two users are alike: Generating audiences with neural clustering for temporal point processes

V Zhuzhel, V Grabar, N Kaploukhaya… - Doklady …, 2023 - Springer
Identifying the right user to target is a common problem for different Internet platforms.
Although numerous systems address this task, they are heavily tailored for specific …

Everyone's Preference Changes Differently: Weighted Multi-Interest Retrieval Model

H Shi, Y Gu, Y Zhou, B Zhao, S Gao, J Zhao - arXiv preprint arXiv …, 2022 - arxiv.org
User embeddings (vectorized representations of a user) are essential in recommendation
systems. Numerous approaches have been proposed to construct a representation for the …

[图书][B] Neural-Symbolic Methods For Neural Architecture Design

H Shi - 2023 - search.proquest.com
Modern neural architectures present strong performance on various tasks, but their abilities
are compromised when faced with the tasks requiring robust reasoning or involving abstract …