School of Engineering Department of Computer Science and Engineering 147 Efficient Queries over Database Supervisor: WONG Raymond Chi Wing / CSE Student: WU Hengxuan / DSCT Course: UROP 1100, Spring Dynamic graphs, where the edge weights change with time, model real-world systems like transportation networks and communication networks. This undergraduate research is interested in finding efficient algorithms for finding the shortest path from a source node (s) to a destination node (t) in a dynamic graph using an oracle or a knowledge data structure. In Milestone 1, we address the problem of determining the optimal time interval on which a given path (P) between vertices (s) and (t) can be traversed with the minimum total weight, which represents the minimum traversal time. With an oracle-based approach being developed and implemented, we ask and compute the time-dependent weights of the edges in an optimal way to achieve the minimum traversal time for the path. Our methodology involves problem definition, design of the oracle’s query mechanism, and calculation of the computational complexity of the solution. Our work can be applied to large-scale dynamic graph routing in the future, with the potential to make major contributions in fields from logistics to network optimization and real-time direction systems. Knowledge Discovery over Database Supervisor: WONG Raymond Chi Wing / CSE Student: CHEN Shr-en / DSCT Course: UROP 1100, Fall Shortest path on a dynamic spatial network has been an important topic in the field of computer science. To answer the dynamic shortest path efficiently, one may want to precompute some results and stored them in an oracle. However, since the graph is dynamic in the sense the weights changes overtime, the oracles have to be updated. Therefore, it’s difficult to balance the trade-off between the precomputation and the update. My project investigate this problem based on the approach in Wei, Wong, and Long (2024). It solves the dynamic shortest path problem efficiently with high probability. In this report, I will present some attempts to improve the result to be deterministically. Knowledge Discovery over Database Supervisor: WONG Raymond Chi Wing / CSE Student: CHEUNG Alan Pak To / COMP Course: UROP 2100, Fall The rapid advancement of Large Language Models (LLMs), exemplified by ChatGPT, Claude, and Gemini has catalyzed the widespread deployment of LLM-based AI agents across various domains. Although AI agents excel in some kinds of tasks such as summarizing and writing, they still exhibit notable limitations in other tasks like arithmetic calculations and logical reasoning. In this UROP project, we explore two innovative and interesting applications of AI agents and investigate three novel prompt engineering methods to enhance AI agents’ reasoning abilities. Additionally, we replace the traditional linear list memory structure with Tree Memory, which can optimize memory organization and response accuracy of AI agents.
RkJQdWJsaXNoZXIy NDk5Njg=