UC Riverside

Name: Amey Bhangale
Title: Finding approximate solutions to computationally hard problems
Description: Numerous practical computational challenges, ranging from scheduling flights and designing chips to network routing and identifying social network communities, can be effectively modeled as discrete optimization problems. The creation of efficient algorithms to address these issues carries significant implications for diverse fields like economics, manufacturing, and engineering. However, the complexity of the majority of these problems is established as NP-hard. This project seeks to overcome this complexity by developing approximation algorithms for a range of discrete optimization problems.
Within this project, the REU student will focus on formulating an approximation algorithm for a specific optimization problem. This task involves obtaining heuristic results through programming in a language of the student's choice.
Preferred qualifications: We're looking for applicants who are naturally curious and passionate about delving into the theoretical side of algorithms and their real-world applications. Good problem-solving skills are important, and if you've taken part in competitive programming contests, that's seen as a positive sign of your knack for algorithms and problem-solving.

Name: Heng Yin
Title: A web application for binary code reverse engineering
Description: This project aims to improve an existing web application for binary code reverse engineering. The students can improve web programming skills and knowledge related to binary code reveres engineering.
Preferred qualifications: Web programing skills; Assembly Code; Computer Security knowledge is preferred.

Name: Philip Brisk
Title: FPGA Application Development
Description: This project will provide training on how to program Field Programmable Gate Arrays (FPGAs) using C/C++ in order to accelerate applications. FPGAs offer significant performance and power consumption advantages compared to CPUs; while they rarely match GPUs in terms of performance, they consume far less power, which makes them an attractive platform to deploy applications, both in the cloud and on the edge.
Preferred qualifications: C/C++ programming skills; experience with VHDL or Verilog is optional, but is not needed.

Name: Philip Brisk
Title: CPU Simulator Development
Description: Companies such as Intel, AMD, ARM, and Nvidia construct detailed software simulators that characterize their next-generation platforms, long before they actually start designing the hardware. This project will teach students how to use simulators to characterize application performance and will provide opportunities to contribute to the development of a first-of-its-kind research simulator that will be used to conduct future research studies.
Preferred qualifications: Interest in Computer Architecture and Computer Systems Organization; C/C++ programming experience; optional: interest/experience with machine learning.

Name: Philip Brisk
Title: Compute-in-Memory
Description: Compute-in-Memory is an emerging form of data storage which can perform a limited set of useful computations (e.g., matrix multiplication). This project will provide students with training on how to program Compute-in-Memory platforms as well as how they are designed and evaluated.
Preferred qualifications: (Helpful) Prior coursework on computer architecture and linear algebra

Name: Yan Gu
Title: Parallelism in Algorithm Design
Description: We have many (~10) ongoing projects and we welcome undergraduate students to join. You are welcome to read my papers to get a better idea: https://www.cs.ucr.edu/~ygu/publications/index.html.
Preferred qualifications: Before we work, you should take our course "CS 214: Parallel Algorithms" online. Feel free to email me and get the course materials.

Name: Yue Dong
Title: Exploring Gender Biases in Large-Scale Language Models
Description: This research project aims to investigate gender biases inherent in large-scale language models. As these models become increasingly prevalent in various sectors—ranging from customer service to decision-making algorithms — the imperative to identify and mitigate biases grows. Our study is structured to dissect the layers of machine learning algorithms, with a keen focus on how gender biases are propagated, perpetuated, and can potentially be rectified in language processing systems.
Preferred qualifications: Python programming skills

Name: Zhaowei Tan
Title: Building Long-Range, Low-Power Internet of Things Systems
Description: LoRa is a wireless communication technology that excels in providing extensive coverage and low power consumption. It is particularly suitable for connecting a multitude of Internet of Things (IoT) devices over long distances in a cost-effective and energy-efficient manner. LoRa operates in unlicensed frequency bands, allowing it to be deployed by various entities without the need for complex regulatory approvals. The technology is often employed in applications that require remote monitoring and control, such as smart agriculture, environmental sensing, asset tracking, and industrial automation.
In our research project, we embark on the journey of creating a functional LoRa system and subsequently assessing its performance. This involves the establishment of crucial components, including real LoRa sensors, gateways, and an IoT cloud server utilizing the popular platform, TheThingsNetwork. Through meticulous configuration, we enable the collection of data from LoRa devices and its visualization in the cloud. Subsequently, comprehensive testing ensues, focusing on validating the system's performance across various dimensions, including its data handling efficiency, security measures, and scalability.
Preferred qualifications: Experience with any programming language is preferred. Basic understanding of computer networks would be a plus, but not mandatory.