2024 Projects
Profit Maximization in Short-Term Financial Instrument Trading with Transformer Value Forecasting
Advisor: Dr. Siadat
More details about the project can be found in the paper and accompanying poster “Profit Maximization in Short-Term Financial Instrument Trading with Transformer Value Forecasting” on the project website. The code for the prototype can be found in the GitHub repository https://github.com/im-keefer/Time-Series-Library.
Lecture Notes |
Transformer-Assisted Bug Detection and Diagnosis in Deep Neural Networks
Advisor: Dr. Wardat
This project explores the use of transformer models for automating bug detection and diagnosis in deep neural networks (DNNs). DNNs are widely used in critical applications like autonomous vehicles and medical diagnosis, where accuracy and reliability are crucial. However, debugging these models is challenging and resource-intensive. Traditional trial-and-error methods are time-consuming and inefficient. To address this, we leveraged transformer models such as PLBART, CodeBERT, and GraphCodeBERT for feature extraction from both faulty and correct models. These features were then used to train deep learning models to classify whether a model contains bugs and to identify the type of bug. Our goal is to create a tool that assists developers in quickly identifying and fixing issues in their AI models, thereby enhancing development efficiency and improving model performance.
Lecture Notes |
Spreading Awareness of Artificial Intelligence Inference Capabilities in User Profiling
Advisor: Dr. Wilson
More details about the project can be found on the project website. The code for the applications developed can be found at https://github.com/amyyu116/social-media-simulation and https://github.com/amyyu116/ai-profiling-report.
Lecture Notes |
Adversarial Attacks and Defenses for Panoptic Perception Models in Autonomous Vehicles
Advisor: Dr. Xu
More details about the project and its poster can be found on the project website.
Lecture Notes |
Enhancing LLM Code Repair Using RAG and Stack Overflow Posts
Advisor: Dr. Wardat
More details about the project can be found in the paper and accompanying poster “Enhancing LLM Code Repair Using RAG and Stack Overflow Posts” on the project website. The prototype code is available at https://github.com/Laboratory-software-Innovation/RAGFix.
Lecture Notes |