Benjamin Moseley: Catalogue data in Spring Semester 2023

Name	Dr. Benjamin Moseley
Name variants	Ben Moseley
Address	ETH AI Center ETH Zürich, OAT X 16 Andreasstrasse 5 8092 Zürich SWITZERLAND
E-mail	benjamin.moseley@ai.ethz.ch
URL	https://benmoseley.blog
Department	Mathematics
Relationship	Lecturer

Number	Title	ECTS	Hours	Lecturers
263-5051-00L	AI Center Projects in Machine Learning Research Last cancellation/deregistration date for this ungraded semester performance: Friday, 17 March 2023! Please note that after that date no deregistration will be accepted and the course will be considered as "fail".	4 credits	2V + 1A	A. Ilic, N. Davoudi, M. El-Assady, F. Engelmann, S. Gashi, T. Kontogianni, A. Marx, B. Moseley, G. Ramponi, X. Shen, M. Sorbaro Sindaci
Abstract	The course will give students an overview of selected topics in advanced machine learning that are currently subjects of active research. The course concludes with a final project.
Learning objective	The overall objective is to give students a concrete idea of what working in contemporary machine learning research is like and inform them about current research performed at ETH. In this course, students will be able to get an overview of current research topics in different specialized areas. In the final project, students will be able to build experience in practical aspects of machine learning research, including research literature, aspects of implementation, and reproducibility challenges.
Content	The course will be structured as sections taught by different postdocs specialized in the relevant fields. Each section will showcase an advanced research topic in machine learning, first introducing it and motivating it in the context of current technological or scientific advancement, then providing practical applications that students can experiment with, ideally with the aim of reproducing a known result in the specific field. A tentative list of topics for this year: - fully supervised 3D scene understanding - weakly supervised 3D scene understanding - causal discovery - biological and artificial neural networks - reinforcement learning - visual text analytics - human-centered AI - representation learning. The last weeks of the course will be reserved for the implementation of the final project. The students will be assigned group projects in one of the presented areas, based on their preferences. The outcomes will be made into a scientific poster and students will be asked to present the projects to the other groups in a joint poster session.
Prerequisites / Notice	Participants should have basic knowledge about machine learning and statistics (e.g. Introduction to Machine Learning course or equivalent) and programming.
401-4656-DRL	Deep Learning in Scientific Computing Only for ETH D-MATH doctoral students and for doctoral students from the Institute of Mathematics at UZH. The latter need to send an email to Jessica Bolsinger (info@zgsm.ch) with the course number. The email should have the subject „Graduate course registration (ETH)“.	1 credit	2V + 1U	S. Mishra, B. Moseley
Abstract	Machine Learning, particularly deep learning is being increasingly applied to perform, enhance and accelerate computer simulations of models in science and engineering. This course aims to present a highly topical selection of themes in the general area of deep learning in scientific computing, with an emphasis on the application of deep learning algorithms for systems, modeled by PDEs.
Learning objective	The objective of this course will be to introduce students to advanced applications of deep learning in scientific computing. The focus will be on the design and implementation of algorithms as well as on the underlying theory that guarantees reliability of the algorithms. We will provide several examples of applications in science and engineering where deep learning based algorithms outperform state of the art methods.
Content	A selection of the following topics will be presented in the lectures. 1. Issues with traditional methods for scientific computing such as Finite Element, Finite Volume etc, particularly for PDE models with high-dimensional state and parameter spaces. 2. Introduction to Deep Learning: Artificial Neural networks, Supervised learning, Stochastic gradient descent algorithms for training, different architectures: Convolutional Neural Networks, Recurrent Neural Networks, ResNets. 3. Theoretical Foundations: Universal approximation properties of the Neural networks, Bias-Variance decomposition, Bounds on approximation and generalization errors. 4. Supervised deep learning for solutions fields and observables of high-dimensional parametric PDEs. Use of low-discrepancy sequences and multi-level training to reduce generalization error. 5. Uncertainty Quantification for PDEs with supervised learning algorithms. 6. Deep Neural Networks as Reduced order models and prediction of solution fields. 7. Active Learning algorithms for PDE constrained optimization. 8. Recurrent Neural Networks and prediction of time series for dynamical systems. 9. Physics Informed Neural networks (PINNs) for the forward problem for PDEs. Applications to high-dimensional PDEs. 10. PINNs for inverse problems for PDEs, parameter identification, optimal control and data assimilation. All the algorithms will be illustrated on a variety of PDEs: diffusion models, Black-Scholes type PDEs from finance, wave equations, Euler and Navier-Stokes equations, hyperbolic systems of conservation laws, Dispersive PDEs among others.
Lecture notes	Lecture notes will be provided at the end of the course.
Literature	All the material in the course is based on research articles written in last 1-2 years. The relevant references will be provided.
Prerequisites / Notice	The students should be familiar with numerical methods for PDEs, for instance in courses such as Numerical Methods for PDEs for CSE, Numerical analysis of Elliptic and Parabolic PDEs, Numerical methods for hyperbolic PDEs, Computational methods for Engineering Applications. Some familiarity with basic concepts in machine learning will be beneficial. The exercises in the course rely on standard machine learning frameworks such as KERAS, TENSORFLOW or PYTORCH. So, competence in Python is helpful.
401-4656-21L	Deep Learning in Scientific Computing Aimed at students in a Master's Programme in Mathematics, Engineering and Physics.	6 credits	2V + 1U	S. Mishra, B. Moseley
Abstract	Machine Learning, particularly deep learning is being increasingly applied to perform, enhance and accelerate computer simulations of models in science and engineering. This course aims to present a highly topical selection of themes in the general area of deep learning in scientific computing, with an emphasis on the application of deep learning algorithms for systems, modeled by PDEs.
Learning objective	The objective of this course will be to introduce students to advanced applications of deep learning in scientific computing. The focus will be on the design and implementation of algorithms as well as on the underlying theory that guarantees reliability of the algorithms. We will provide several examples of applications in science and engineering where deep learning based algorithms outperform state of the art methods.
Content	A selection of the following topics will be presented in the lectures. 1. Issues with traditional methods for scientific computing such as Finite Element, Finite Volume etc, particularly for PDE models with high-dimensional state and parameter spaces. 2. Introduction to Deep Learning: Artificial Neural networks, Supervised learning, Stochastic gradient descent algorithms for training, different architectures: Convolutional Neural Networks, Recurrent Neural Networks, ResNets. 3. Theoretical Foundations: Universal approximation properties of the Neural networks, Bias-Variance decomposition, Bounds on approximation and generalization errors. 4. Supervised deep learning for solutions fields and observables of high-dimensional parametric PDEs. Use of low-discrepancy sequences and multi-level training to reduce generalization error. 5. Uncertainty Quantification for PDEs with supervised learning algorithms. 6. Deep Neural Networks as Reduced order models and prediction of solution fields. 7. Active Learning algorithms for PDE constrained optimization. 8. Recurrent Neural Networks and prediction of time series for dynamical systems. 9. Physics Informed Neural networks (PINNs) for the forward problem for PDEs. Applications to high-dimensional PDEs. 10. PINNs for inverse problems for PDEs, parameter identification, optimal control and data assimilation. All the algorithms will be illustrated on a variety of PDEs: diffusion models, Black-Scholes type PDEs from finance, wave equations, Euler and Navier-Stokes equations, hyperbolic systems of conservation laws, Dispersive PDEs among others.
Lecture notes	Lecture notes will be provided at the end of the course.
Literature	All the material in the course is based on research articles written in last 1-2 years. The relevant references will be provided.
Prerequisites / Notice	The students should be familiar with numerical methods for PDEs, for instance in courses such as Numerical Methods for PDEs for CSE, Numerical analysis of Elliptic and Parabolic PDEs, Numerical methods for hyperbolic PDEs, Computational methods for Engineering Applications. Some familiarity with basic concepts in machine learning will be beneficial. The exercises in the course rely on standard machine learning frameworks such as KERAS, TENSORFLOW or PYTORCH. So, competence in Python is helpful.