View Proposal


Proposer
Alessandro Suglia
Title
Multi-agent AI Systems using LLM
Goal
Using LLMs to develop Multi-agent Learning systems
Description
Traditional Natural Language Processing systems have been designed with a single task in mind such as sentiment analysis, text classification, etc. Thanks to Large Language Models like ChatGPT, these AI systems can now solve multiple tasks depending on the text prompt that they receive. However, a single AI agent wouldn't be sufficient to solve very complex and multi-step tasks required by intricate real-world applications. In this project, you will work with the most recent LLMs and combine them in novel ways to create complex multi-agent applications.
Resources
Microsoft AutoGen: https://microsoft.github.io/autogen/0.2/ OpenAI Swarms: https://github.com/openai/swarm
Background
https://arxiv.org/abs/2402.01680
Url
Difficulty Level
Challenging
Ethical Approval
None
Number Of Students
2
Supervisor
Alessandro Suglia
Keywords
multi-agent systems,llms,generative ai,deep learning
Degrees
Master of Science in Artificial Intelligence
Master of Science in Human Robot Interaction
Master of Science in Robotics