学术讲座

大数据管理与分析方法研究北京市重点实验室（BDAI）研究生沙龙由中国人民大学高瓴人工智能智能学院与信息学院联合定期举行，本周BDAI重点实验室研讨会由高瓴人工智能学院博士生袁深和信息学院博士生孙路明分别介绍各自的研究工作。欢迎同学们积极参与研讨！

汇报人：袁深，博士一年级

时间：2021年11月10日 12：30-13：30

标题：Self-Organized Hawkes Processes

摘要：In this paper, we propose a novel self-organized Hawkes process (SOHP) to model complex event sequences based on extremely few observations. Motivated by the fact that the complicated global relations among events are often composed of simple local relations, we model the event sequences by a set of heterogeneous local Hawkes processes rather than a single Hawkes process. In the training phase, we learn the Hawkes processes with a self-organization mechanism, selecting training sequences adaptively for each Hawkes process by a bandit algorithm. The reward used in the algorithm is originally defined based on an optimal transport distance. Additionally, we leverage the superposition property of the Hawkes process to enhance the robustness of our algorithm to the data sparsity problem. We apply our SOHP method to sequential recommendation problems in the continuous-time domain and achieve encouraging performance in various datasets.

汇报人：孙路明，博士三年级

时间：2021年11月10日 12：30-13：30

标题：AI Meets Query Optimization

摘要：Traditional database systems and data management techniques are facing great challenge due to the 3V’s in Big Data. The development of artificial intelligence provides a brand-new opportunity for database management systems with its power in learning, reasoning and planning. Through learning from data distribution, query workload and query execution performance, the systems powered by artificial intelligence are able to forecast future workload, tune database configurations, partition data blocks, index o n proper columns, estimate selectivity, optimize query plan and control query concurrency automatically. Also, some machine learning models can replace core components of a database such as index structures. We introduce new research on database systems with artificial intelligence and state the existing problems and potential solutions. And we propose MOSE, a learning-based MOnotonic Selectivity Estimator for query optimization. Mose can offer accurate and fast selectivity estimation for different dataset and workload, maintaining basic rules in selectivity estimation such as monotonicity and consistency.