Solving Differential Equations using Neural Networks
Project Level: Summer
Project Duration: 6 weeks
Hours of Engagement: 20-36 hours
Project Description:
“Science is a differential equation.” – Alan Turing
Differential equations offer a powerful modelling tool for understanding our world. They have diverse applications in domains including fluid flow, electromagnetism, epidemiology. However, many differential equations are difficult to solver, whether analytically or numerically. Neural networks have recently been shown to be promising efficient approximate solutions. This project will explore some ideas in this direction.
Expected Outcomes:
• Develop skills for implementing neural network for solving differential equations.
• Develop skills in using existing tools
• Develop skills in research design, implementation, experimentation, and communication.
• A report documenting the work done and the findings.
Suitable for:
Essential: knowledge of differential equations and neural networks
Desirable: knowledge of numerical methods for solving differential equations
Contact for further information:
Dr Nan Ye: nan.ye@uq.edu.au