Lecture / Seminar
일 일 일 Sun | 월 월 월 Mon | 화 화 화 Tue | 수 수 수 Wed | 목 목 목 Thu | 금 금 금 Fri | 토 토 토 Sat |
---|---|---|---|---|---|---|
MINDS Seminar on Machine Learning
분야Field | |||
---|---|---|---|
날짜Date | 2024-04-23 ~ 2024-04-23 | 시간Time | 10:00 ~ 11:00 |
장소Place | online | 초청자Host | |
연사Speaker | Jihun Han | 소속Affiliation | Dartmouth College |
TOPIC | MINDS Seminar on Machine Learning | ||
소개 및 안내사항Content | Title : Collaboration between PDEs and machine learning: solving PDEs and learning dynamical systems using neural networks Speaker : Jihun Han (Dartmouth College) Abstract : This talk has two main parts. Firstly, I will discuss a stochastic approach for solving PDEs: the derivative-free loss method (DFLM). I will cover its analysis and highlight its advantages in addressing multiscale problems and perforated domain problems in a non-intrusive manner. In the second part, I will introduce a method for learning in-between imagery dynamics. This method involves integrating PDE models within the latent spaces to enhance learning. Notably, this method exhibits robustness in capturing intricate dynamics, such as rotation and outflow, which are challenging for existing state-of-the-art methods of optimal transport. |
학회명Field | MINDS Seminar on Machine Learning | ||
---|---|---|---|
날짜Date | 2024-04-23 ~ 2024-04-23 | 시간Time | 10:00 ~ 11:00 |
장소Place | online | 초청자Host | |
소개 및 안내사항Content | Title : Collaboration between PDEs and machine learning: solving PDEs and learning dynamical systems using neural networks Speaker : Jihun Han (Dartmouth College) Abstract : This talk has two main parts. Firstly, I will discuss a stochastic approach for solving PDEs: the derivative-free loss method (DFLM). I will cover its analysis and highlight its advantages in addressing multiscale problems and perforated domain problems in a non-intrusive manner. In the second part, I will introduce a method for learning in-between imagery dynamics. This method involves integrating PDE models within the latent spaces to enhance learning. Notably, this method exhibits robustness in capturing intricate dynamics, such as rotation and outflow, which are challenging for existing state-of-the-art methods of optimal transport. |
성명Field | MINDS Seminar on Machine Learning | ||
---|---|---|---|
날짜Date | 2024-04-23 ~ 2024-04-23 | 시간Time | 10:00 ~ 11:00 |
소속Affiliation | Dartmouth College | 초청자Host | |
소개 및 안내사항Content | Title : Collaboration between PDEs and machine learning: solving PDEs and learning dynamical systems using neural networks Speaker : Jihun Han (Dartmouth College) Abstract : This talk has two main parts. Firstly, I will discuss a stochastic approach for solving PDEs: the derivative-free loss method (DFLM). I will cover its analysis and highlight its advantages in addressing multiscale problems and perforated domain problems in a non-intrusive manner. In the second part, I will introduce a method for learning in-between imagery dynamics. This method involves integrating PDE models within the latent spaces to enhance learning. Notably, this method exhibits robustness in capturing intricate dynamics, such as rotation and outflow, which are challenging for existing state-of-the-art methods of optimal transport. |
성명Field | MINDS Seminar on Machine Learning | ||
---|---|---|---|
날짜Date | 2024-04-23 ~ 2024-04-23 | 시간Time | 10:00 ~ 11:00 |
호실Host | 인원수Affiliation | Jihun Han | |
사용목적Affiliation | 신청방식Host | Dartmouth College | |
소개 및 안내사항Content | Title : Collaboration between PDEs and machine learning: solving PDEs and learning dynamical systems using neural networks Speaker : Jihun Han (Dartmouth College) Abstract : This talk has two main parts. Firstly, I will discuss a stochastic approach for solving PDEs: the derivative-free loss method (DFLM). I will cover its analysis and highlight its advantages in addressing multiscale problems and perforated domain problems in a non-intrusive manner. In the second part, I will introduce a method for learning in-between imagery dynamics. This method involves integrating PDE models within the latent spaces to enhance learning. Notably, this method exhibits robustness in capturing intricate dynamics, such as rotation and outflow, which are challenging for existing state-of-the-art methods of optimal transport. |