AAAI 2023 | 推荐系统，多任务学习，因果推断相关论文整理

AAAI  2023的论文已经出来了，笔者整理了推荐，多任务和因果推断相关的一些文章，方便大家查阅。值得注意的是，因果推断仍然是热点。

查阅所有论文可前往链接：https://aaai-23.aaai.org/wp-content/uploads/2023/01/Updated_AAAI-23-Technical-Schedule-FEB08.pdf

推荐系统

Fair Representation Learning for Recommendation: A Mutual Information Perspective 【推荐的公平表示学习：互信息的视角】

Untargeted Attack against Federated Recommendation Systems via Poisonous Item Embeddings and the Defense【通过有毒项目嵌入和防御对联邦推荐系统的无目标攻击】

PPGenCDR: A Stable and Robust Framework for Privacy-Preserving Cross-domain Recommendation 【PPGenCDR：一个稳定而鲁棒的隐私保护跨域推荐框架】

Adaptive Low-Precision Training for Embeddings in Click-Through Rate Prediction【点击率预测中嵌入的自适应低精度训练】

Factual and Informative Review Generation for Explainable Recommendation 【可解释推荐的事实和信息审查生成】

Cross-domain Adaptative Learning for Online Advertisement Customer Lifetime Value Prediction 【在线广告客户终身价值预测的跨域自适应学习】

Uniform Sequence Better: Time Interval Aware Data Augmentation for Sequential Recommendation 【更好的统一序列：用于序列推荐的时间间隔感知数据增强】

LANCER: A Lifetime-Aware News Recommender System 【LANCER：新闻推荐系统】

Context-aware Safe Medication Recommendations with Molecular Graph and DDI Graph Embedding【基于分子图和DDI图嵌入的上下文感知安全用药建议】  

CP-Rec: Contextual Prompting for Conversational Recommender Systems 【CP Rec:会话推荐系统的上下文提示】

Multiple Robust Learning for Recommendation  【推荐的多重鲁棒学习】

Win-Win: A Privacy-Preserving Federated Framework for DualTarget Cross-Domain Recommendation 【双赢：双目标跨域推荐的隐私保护联邦框架】

Towards Reliable Item Sampling for Recommendation Evaluation  【推荐评估的可靠项目抽样】

Structure Aware Incremental Learning with Personalized Imitation Weights for Recommender Systems【基于个性化模仿权重的推荐系统结构感知增量学习】  

CowClip: Reducing CTR Prediction Model Training Time from 12 hours to 10 minutes on 1 GPU 【CowClip：在1个GPU上将CTR预测模型训练时间从12小时减少到10分钟】

多任务

AdaTask: A Task-aware Adaptive Learning Rate Approach to Multitask Learning 【AdaTask：一种多任务学习的任务感知自适应学习率方法】

PiCor: Multi-Task Deep Reinforcement Learning with Policy Correction 【PiCor:多任务深度强化学习与策略修正】

Learning Conflict-Noticed Architecture for Multi-Task Learning 【基于学习冲突的多任务学习体系结构】

MIGA: A Unified Multi-task Generation Framework for Conversational Text-to-SQL  【MIGA：一个用于会话文本到SQL的统一多任务生成框架】

因果推断

Learning Instrumental Variable from Data Fusion for Treatment Effect Estimation 【用于干预效果评估的数据融合学习工具变量】

Knowledge-Bridged Causal Interaction Network for Causal Emotion Entailment 【基于知识桥的因果互动网络】

Estimating average causal effects from patient trajectories 【根据患者轨迹估计平均因果效应】

Disentangled Representation for Causal Mediation Analysis 【因果调解分析的离散表示】

Causal Inference with Conditional Instruments using Deep Generative Models 【使用深度生成模型的条件工具因果推断】

Robust Causal Graph Representation Learning against Confounding Effects 【抗混淆效应的鲁棒因果图表示学习】

Causal Recurrent Variational Autoencoder for Medical Time Series Generation 【用于医学时间序列生成的因果递归变分自动编码器】

Improvement-Focused Causal Recourse (ICR) 【基于改进的因果关系追索权（ICR）】

COCA: COllaborative CAusal Regularization for Audio-Visual Question Answering 【COCA：用于视听问答的协作因果规则化】

Direct Heterogeneous Causal Learning for Resource Allocation Problems in Marketing  【市场营销中资源分配问题的直接异质因果学习】

Self-supervised Learning for Multilevel Skeleton-based Forgery Detection via Temporal-Causal Consistency of Actions  【基于时间因果一致性的多级骨架伪造检测的自监督学习】

Causal Effect Identification in Cluster DAGs  【聚类DAG中因果关系的识别】

Learning Relational Causal Models with Cycles through Relational Acyclification 【通过关系非循环化学习具有循环的关系因果模型】

Weather2vec: Representation Learning for Causal Inference with Non-Local Confounding in Air Pollution and Climate Studies 【Weather2vec：空气污染和气候研究中非局部混淆因果推断的表征学习】

Counterfactual Dynamics Forecasting - A New Setting of Quantitative Reasoning  【反事实动力学预测——一种新的定量推理方法】

Video-Audio Domain Generalization via Confounder Disentanglement 【基于混淆因子解耦的视频音频域泛化】

Formalising the Robustness of Counterfactual Explanations for Neural Networks  【神经网络反事实解释的鲁棒性形式化】

Very Fast, Approximate Counterfactual Explanations for Decision Forests 【决策森林的快速近似反事实解释】

Counterfactual Learning with General Data-generating Policies 【反事实学习与一般数据生成策略】

一起交流

想和你一起学习进步！『NewBeeNLP』目前已经建立了多个不同方向交流群（机器学习 / 深度学习 / 自然语言处理 / 搜索推荐 / 图网络 / 面试交流 / 等），名额有限，赶紧添加下方微信加入一起讨论交流吧！（注意一定o要备注信息才能通过）