DeepStock: Reinforcement Learning with Policy Regularizations for Inventory Management

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xie, Yaqi, Hao, Xinru, Liu, Jiaxi, Ma, Will, Xin, Linwei, Cao, Lei, Zhang, Yidong
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910060752404480
author Xie, Yaqi
Hao, Xinru
Liu, Jiaxi
Ma, Will
Xin, Linwei
Cao, Lei
Zhang, Yidong
author_facet Xie, Yaqi
Hao, Xinru
Liu, Jiaxi
Ma, Will
Xin, Linwei
Cao, Lei
Zhang, Yidong
contents Deep Reinforcement Learning (DRL) provides a general-purpose methodology for training inventory policies that can leverage big data and compute. However, off-the-shelf implementations of DRL have seen mixed success, often plagued by high sensitivity to the hyperparameters used during training. In this paper, we show that by imposing policy regularizations, grounded in classical inventory concepts such as "Base Stock", we can significantly accelerate hyperparameter tuning and improve the final performance of several DRL methods. We report details from a 100% deployment of DRL with policy regularizations on Alibaba's e-commerce platform, Tmall. We also include extensive synthetic experiments, which show that policy regularizations reshape the narrative on what is the best DRL method for inventory management.
format Preprint
id arxiv_https___arxiv_org_abs_2603_19621
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle DeepStock: Reinforcement Learning with Policy Regularizations for Inventory Management
Xie, Yaqi
Hao, Xinru
Liu, Jiaxi
Ma, Will
Xin, Linwei
Cao, Lei
Zhang, Yidong
Machine Learning
Artificial Intelligence
Deep Reinforcement Learning (DRL) provides a general-purpose methodology for training inventory policies that can leverage big data and compute. However, off-the-shelf implementations of DRL have seen mixed success, often plagued by high sensitivity to the hyperparameters used during training. In this paper, we show that by imposing policy regularizations, grounded in classical inventory concepts such as "Base Stock", we can significantly accelerate hyperparameter tuning and improve the final performance of several DRL methods. We report details from a 100% deployment of DRL with policy regularizations on Alibaba's e-commerce platform, Tmall. We also include extensive synthetic experiments, which show that policy regularizations reshape the narrative on what is the best DRL method for inventory management.
title DeepStock: Reinforcement Learning with Policy Regularizations for Inventory Management
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2603.19621