Weipeng P. Yan

Learn More
Efficient processing of aggregation queries is essential for decision support applications. This paper describes a class of query transformations, called eager aggregation and laty aggregation, that allows a query optimizer to move group-by operations up and down the query tree. Eager aggregation partially pushes a groupby past a join. After a group-by is(More)
Assume that we have an SQL query containing joins and a group by The standard way of evaluating this type of query is to rst perform all the joins and then the group by operation However it may be possible to perform the group by early that is to push the group by operation past one or more joins Early grouping may reduce the query processing cost by(More)
SQL queries containing GROUP BY and aggre-gation occur frequently in decision support applications. Grouping with aggregation is typically done by rst sorting the input and then performing the aggregation as part of the output phase of the sort. The most widely used external sorting algorithm is merge sort, consisting of a run formation phase followed by a(More)
We present LADDER, the first deep reinforcement learning agent that can successfully learn control policies for largescale real-world problems directly from raw inputs composed of high-level semantic information. The agent is based on an asynchronous stochastic variant of DQN (Deep Q Network) named DASQN. The inputs of the agent are plain-text descriptions(More)
Designing an e-commerce recommender system that serves hundreds of millions of active users is a daunting challenge. Ranking strategy as the key module needs to be more carefully designed. We find two key factors that affect users’ behaviors: attractive item content and compatibility with users’ interests. To extract these factors, a ranking model needs to(More)
With the transition from people’s traditional ‘brickand-mortar’ shopping to online mobile shopping patterns in web 2.0 era, the recommender system plays a critical role in E-Commerce and E-Retails. This is especially true when designing this system for more than 236 million active users. Ranking strategy, the key module of the recommender system, needs to(More)
  • 1