layout | permalink | title | header | ||||
---|---|---|---|---|---|---|---|
splash |
/ |
Compute Fest 2024 : PINNs Course |
|
Teaching Assistants: Shuheng Liu, Kshitij Parwani, Wanzhou Lei , Lakshay Chawla , Sathvik Bhagavan
Welcome to the Course on Physics-Informed Neural Networks (PINNs)
We are thrilled to present this course as one of the very few comprehensive programs on Physics-Informed Neural Networks (PINNs). Carefully curated, this course serves as a vital link between the traditional field of differential equations and the rapidly evolving discipline of neural networks. It is optimally designed for those who possess a foundational understanding of machine learning and are keen to apply this knowledge to differential equations.
The course features live lectures, hands-on coding exercises, and self-assessment quizzes to offer a well-rounded educational experience.
Our dedicated team of educators stands ready to provide guidance and support, addressing your questions both through our specialized course discussion forum and during live online office hours.
- Introduction to Differential Equations: Basic concepts, classifications, and real-world applications
- Numerical Methods for DE: Techniques to solve differential equations using computational algorithms
- Neural Networks and DE: Exploring NN-based methods for solving differential equations
- Setting Initial Conditions: Handling initial and boundary values
- Optimization and Sampling: Dive into methods of optimization and various sampling techniques
- Transfer Learning for PINNs: Understand the application of transfer learning specifically in the PINNs realm
- Error Bounds and Uncertainty: Grasp the concepts of error bounds and how to quantify uncertainties in PINNs
It would be helpful to have a foundational understanding of differential equations and a basic grasp of neural network concepts. This course is tailored for those looking to refresh their understanding of advanced topics or to fill any gaps in prior knowledge. Those entirely new to differential equations or neural networks may find the pace a tad challenging, but remember, our support team is here to guide you every step of the way.
The learning materials for each topic are bundled. Lessons will help students develop the intuition for core concepts, providing the necessary mathematical background, guidance on technical details, and relevant examples.
Lessons include readings, lecture slides, and lecture videos. The goal of the reading is to prepare you for the lesson content, familiarize you with new terminology and definitions, and to help you determine which part of the subject may need more attention.
Each session may have a short 'reading check' at the beginning which covers the assigned reading for that week. This will help assess your understanding of the material.
Most lessons also include exercises and quizzes. Quizzes are for self-assessment. The exercises can be attempted as many times as you wish.
For questions about the course content, after you have tried to troubleshoot on your own, the process to get help is as follows:
- Post your question on the course discussion forum. Note that forum posts are visible to everyone, and the teaching staff monitors these posts. You are encouraged to answer your peers' questions! 😁
- Attend an online office hour (schedule TBD)."
We actively seek and welcome people of diverse identities, from across the spectrum of disciplines and methods since Artificial Intelligence (AI) increasingly mediates our social, cultural, economic, and political interactions.
We believe in creating and maintaining an inclusive learning environment where all members feel safe, respected, and capable of producing their best work.
We commit to an experience for all participants that is free from -- Harassment, bullying, and discrimination which includes but is not limited to:
- Offensive comments related to age, race, religion, creed, color, gender (including transgender/gender identity/gender expression), sexual orientation, medical condition, physical or intellectual disability, pregnancy, or medical conditions, national origin or ancestry.
- Intimidation, personal attacks, harassment, unnecessary disruption of talks during any of the learning activities.
Scope of Permissible Utilization: These Educational Resources are primarily provided for the academic enrichment of students registered in the course. Others may access these resources solely for individual educational purposes.
-
Commercial exploitation or sale of any part of the Educational Resources is categorically prohibited.
-
Redistribution, dissemination, or publication of any segment of the Educational Resources, in any format or through any medium, is strictly forbidden without prior explicit permission from the course instructor.
-
Alteration, modification, or the creation of derivative works from the Educational Resources is not authorized without prior permission.