View Proposal

---

Proposer

Claudio Zito

Title

Reinforcement Learning for Autonomous Agents

Goal

To investigate and implement reinforcement learning (RL) techniques for developing autonomous agents that can learn and adapt their decision-making strategies in dynamic environments, enhancing their ability to interact and perform tasks effectively.

Description

This project focuses on the application of reinforcement learning methods to train autonomous agents capable of adapting to varying conditions and objectives. Students will explore various RL algorithms, such as Q-learning, Deep Q-Networks (DQN), and Policy Gradients, to develop agents that can learn from their experiences and optimize their actions based on reward feedback. The project allows for flexibility in application, enabling students to select specific tasks, environments, and learning objectives for their autonomous agents. Advanced students will explore how LLM can be used as agentic RL frameworks.

Resources

Hardware: Computing resources (laptops or desktops with GPUs for training), robotic simulation environments (e.g., Unity, OpenAI Gym, or custom simulation). Software: RL libraries (e.g., Stable Baselines3, TensorFlow, or PyTorch) for implementing algorithms Simulation platforms compatible with ROS for integrating with robotic environments Visualization tools for monitoring agent performance and behavior. Documentation and Tutorials: Online resources for reinforcement learning algorithms, articles defining autonomous agents, and simulation platform guides.

Background

Reinforcement learning (RL) is a powerful approach in artificial intelligence that allows agents to learn optimal behaviors through trial and error by receiving feedback in the form of rewards or penalties. This field has significant implications for autonomous systems, including robotics, gaming, and autonomous vehicles. By investigating RL for autonomous agents, this project aims to enhance their ability to operate in complex and unpredictable environments, paving the way for more intelligent and capable systems.

Url

External Link

Difficulty Level

Easy

Ethical Approval

None

Number Of Students

0

Supervisor

Claudio Zito

Keywords

reinforcement learning (rl), autonomous agents, decision-making strategies, q-learning, policy gradient methods, simulation environments, adaptive learning, trial and error, multi-agent systems, exploration vs. exploitation

Degrees

Bachelor of Science in Computer Science
Bachelor of Science in Computer Systems
Bachelor of Science in Information Systems
Master of Science in Artificial Intelligence
Master of Science in Artificial Intelligence with SMI
Master of Science in Computing (2 Years)
Master of Science in Data Science
Master of Science in Human Robot Interaction
Master of Science in Information Technology (Business)
Master of Science in Information Technology (Software Systems)
Master of Science in Robotics
Master of Science in Software Engineering
Bachelor of Science in Computing Science
Bachelor of Engineering in Robotics
Master of Science in Robotics with Industrial Application
Postgraduate Diploma in Artificial Intelligence
Bachelor of Science in Statistical Data Science
BSc Data Sciences