Deep Reinforcement Learning for Page-wise Recommendations

Abstract

Recommender systems can mitigate the information overload problem by suggesting users' personalized items. In real-world recommendations such as e-commerce, a typical interaction between the system and its users is -- users are recommended a page of items and provide feedback; and then the system recommends a new page of items. To effectively capture such interaction for recommendations, we need to solve two key problems -- (1) how to update recommending strategy according to user's \textit{real-time feedback}, and 2) how to generate a page of items with proper display, which pose tremendous challenges to traditional recommender systems. In this paper, we study the problem of page-wise recommendations aiming to address aforementioned two challenges simultaneously. In particular, we propose a principled approach to jointly generate a set of complementary items and the corresponding strategy to display them in a 2-D page; and propose a novel page-wise recommendation framework based on deep reinforcement learning, DeepPage, which can optimize a page of items with proper display based on real-time feedback from users. The experimental results based on a real-world e-commerce dataset demonstrate the effectiveness of the proposed framework.

arXiv admin note: text overlap with arXiv:1802.06501

References in corpus (10)

• Neural Machine Translation by Jointly Learning to Align and Translate
• Continuous control with deep reinforcement learning
• Playing Atari with Deep Reinforcement Learning
• Empirical Analysis of Predictive Algorithms for Collaborative Filtering
• Session-based Recommendations with Recurrent Neural Networks
• Recommendations with Negative Feedback via Pairwise Deep Reinforcement Learning
• PEGASUS: A Policy Search Method for Large MDPs and POMDPs
• Deep Reinforcement Learning for List-wise Recommendations
• Reinforcement Learning based Recommender System using Biclustering Technique
• Deep Reinforcement Learning with Attention for Slate Markov Decision Processes with High-Dimensional States and Actions

Cited by in corpus (84)

• Deep Learning based Recommender System: A Survey and New Perspectives
• Recommendations with Negative Feedback via Pairwise Deep Reinforcement Learning
• Fast Parallel Hypertree Decompositions in Logarithmic Recursion Depth
• Towards Long-term Fairness in Recommendation
• Neural Interactive Collaborative Filtering
• Deep Reinforcement Learning for List-wise Recommendations
• Blockchain-based Recommender Systems: Applications, Challenges and Future Opportunities
• LinRec: Linear Attention Mechanism for Long-term Sequential Recommender Systems
• Horizon: Facebook's Open Source Applied Reinforcement Learning Platform
• Deep reinforcement learning for search, recommendation, and online advertising: a survey
• Jointly Learning to Recommend and Advertise
• A Survey on Session-based Recommender Systems
• HAMUR: Hyper Adapter for Multi-Domain Recommendation
• Multi-Task Fusion via Reinforcement Learning for Long-Term User Satisfaction in Recommender Systems
• RecSim: A Configurable Simulation Platform for Recommender Systems
• Whole-Chain Recommendations
• AutoDenoise: Automatic Data Instance Denoising for Recommendations
• Multi-Task Recommendations with Reinforcement Learning
• Reinforcement Learning Applications
• Knowledge-guided Deep Reinforcement Learning for Interactive Recommendation
• Exploration and Regularization of the Latent Action Space in Recommendation
• Adaptive Reward-Poisoning Attacks against Reinforcement Learning
• Large Language Models are Learnable Planners for Long-Term Recommendation
• User Retention-oriented Recommendation with Decision Transformer
• Modeling Users' Contextualized Page-wise Feedback for Click-Through Rate Prediction in E-commerce Search
• Reinforcement Learning for Slate-based Recommender Systems: A Tractable Decomposition and Practical Methodology
• Sequential Recommendation for Optimizing Both Immediate Feedback and Long-term Retention
• Reinforcement Learning to Optimize Long-term User Engagement in Recommender Systems
• User Tampering in Reinforcement Learning Recommender Systems
• A Survey of Deep Reinforcement Learning in Recommender Systems: A Systematic Review and Future Directions
• Value Penalized Q-Learning for Recommender Systems
• Toward Simulating Environments in Reinforcement Learning Based Recommendations
• Value-aware Recommendation based on Reinforced Profit Maximization in E-commerce Systems
• Provably Efficient Black-Box Action Poisoning Attacks Against Reinforcement Learning
• State Encoders in Reinforcement Learning for Recommendation: A Reproducibility Study
• Top-K Off-Policy Correction for a REINFORCE Recommender System
• Large-scale Interactive Recommendation with Tree-structured Policy Gradient
• Future Impact Decomposition in Request-level Recommendations
• Why People Skip Music? On Predicting Music Skips using Deep Reinforcement Learning
• Learning to Collaborate in Multi-Module Recommendation via Multi-Agent Reinforcement Learning without Communication
• RecSim NG: Toward Principled Uncertainty Modeling for Recommender Ecosystems
• Batch-Constrained Distributional Reinforcement Learning for Session-based Recommendation
• Sequential Evaluation and Generation Framework for Combinatorial Recommender System
• ROLeR: Effective Reward Shaping in Offline Reinforcement Learning for Recommender Systems
• Understanding the Limits of Poisoning Attacks in Episodic Reinforcement Learning
• AutoAssign+: Automatic Shared Embedding Assignment in Streaming Recommendation
• Modeling User Retention through Generative Flow Networks
• Towards Off-Policy Reinforcement Learning for Ranking Policies with Human Feedback
• AutoLoss: Automated Loss Function Search in Recommendations
• MARL with General Utilities via Decentralized Shadow Reward Actor-Critic
• Balancing Accuracy and Fairness for Interactive Recommendation with Reinforcement Learning
• Towards Validating Long-Term User Feedbacks in Interactive Recommendation Systems
• Minimizing Live Experiments in Recommender Systems: User Simulation to Evaluate Preference Elicitation Policies
• AliExpress Learning-To-Rank: Maximizing Online Model Performance without Going Online
• Deep Reinforcement Learning based Group Recommender System
• Reward Poisoning in Reinforcement Learning: Attacks Against Unknown Learners in Unknown Environments
• Multi-task Offline Reinforcement Learning for Online Advertising in Recommender Systems
• Learning to Recommend via Meta Parameter Partition
• Deep Bayesian Bandits: Exploring in Online Personalized Recommendations
• Reinforcement Re-ranking with 2D Grid-based Recommendation Panels
• When Collaborative Filtering Meets Reinforcement Learning
• Interactive Recommender System via Knowledge Graph-enhanced Reinforcement Learning
• Accelerating Offline Reinforcement Learning Application in Real-Time Bidding and Recommendation: Potential Use of Simulation
• Generator and Critic: A Deep Reinforcement Learning Approach for Slate Re-ranking in E-commerce
• Sequential Search with Off-Policy Reinforcement Learning
• Developing Multi-Task Recommendations with Long-Term Rewards via Policy Distilled Reinforcement Learning
• Prompt Tuning as User Inherent Profile Inference Machine
• SPARK: Adaptive Low-Rank Knowledge Graph Modeling in Hybrid Geometric Spaces for Recommendation
• Interpretable performance analysis towards offline reinforcement learning: A dataset perspective
• Empowering Denoising Sequential Recommendation with Large Language Model Embeddings
• Deep Hierarchical Reinforcement Learning Based Recommendations via Multi-goals Abstraction
• Off-policy Learning for Multiple Loggers
• Optimized Recommender Systems with Deep Reinforcement Learning
• A Load Balanced Recommendation Approach
• Value Function Decomposition in Markov Recommendation Process
• Jointly Learning Explainable Rules for Recommendation with Knowledge Graph
• Model-free Reinforcement Learning with Stochastic Reward Stabilization for Recommender Systems
• D2RLIR : an improved and diversified ranking function in interactive recommendation systems based on deep reinforcement learning
• Fast Distributed Bandits for Online Recommendation Systems
• MBCAL: Sample Efficient and Variance Reduced Reinforcement Learning for Recommender Systems
• RLINK: Deep Reinforcement Learning for User Identity Linkage
• Model-free Policy Learning with Reward Gradients
• Locality-Sensitive Experience Replay for Online Recommendation
• Adaptive Neural Architectures for Recommender Systems