Generative methods in Machine learning: Applications in science and engineering
| Dates: | 21 February 2023 |
|---|
| Times: | 15:00 - 17:00 |
|---|
| What is it: | Lecture |
|---|
| Organiser: | Department of Mechanical, Aerospace and Civil Engineering |
|---|
| Who is it for: | University staff, External researchers, Adults, Alumni, General public |
|---|
| Speaker: | Dr. Alex Skillen |
|---|
|
|
Note: this is a hybrid learning event. It is most beneficial to attend in-person
Learning objectives:
1. To provide the audience with an accessible overview of the theory of generative methods; what they are, and when they might be useful in science and engineering. Specifically, we will "get under the hood" of Generative Adversarial networks, Diffusion models, and Transformers, demystifying these complex models for a general audience.
2. To provide a brief literature review of the use in generative methods in fluid mechanics (noting that the methods are transferable to other areas in science and engineering).
Description:
Accurate simulations are often infeasible, while feasible simulations are often inaccurate. This lecture will outline the current literature in addressing this dilemma via data-driven approaches. We will consider generative methods; a class of machine learning that aims to learn the underlying distribution given data. Specifically, we will focus on Generative Adversarial Networks, Diffusion models, and transformers, giving an overview of the theory behind these modern generative approaches. Following a review of current state-of-the-art in generative modelling for fluid flows, we will consider the latest trends and opportunities in the field. Note that while the focus of this review is on generative methods for fluid flows, this is driven by the background of the lecturer. The methods discussed are transferable to other areas such as solid mechanics.
Bio of speakers:
Dr. Alex Skillen is a Lecturer in Engineering Simulation and Data Science within the Department of Mechanical, Aerospace and Civil Engineering at the University of Manchester. Alex’s research interests lie at the intersection of Computational Fluid Dynamics (CFD) and machine learning.
Speaker
Dr. Alex Skillen
Role: Lecturer in Engineering Simulation and Data Science
Organisation: University of Manchester
Travel and Contact Information
Find event
1A.023
Engineering Building A
Greater Manchester
