Computational Mechanics & Artificial Intelligence Lab
Nick Napoleon Vlassis is an Assistant Professor in the Department of Mechanical & Aerospace Engineering at Rutgers University.
Home
About
Publications
Google Scholar
Hiring
Teaching
News Feed
Geometric Deep Learning for Computational Mechanics Part I: Anisotropic Hyperelasticity
Geometric Deep Learning for Computational Mechanics Part II: Graph Embedding for Interpretable Multiscale Plasticity
Sobolev Training of Thermodynamic-Informed Neural Networks for Interpretable Elasto-Plasticity Models with Level Set Hardening
Molecular Dynamics Inferred Transfer Learning Models for Finite-Strain Hyperelasticity of Monoclinic Crystals: Sobolev Training and Validations Against Physical Constraints
Component-Based Machine Learning Paradigm for Discovering Rate-Dependent and Pressure-Sensitive Level-Set Plasticity Models
Denoising diffusion algorithm for inverse design of microstructures with fine-tuned nonlinear material properties
nick.vlassis@rutgers.edu
16:650:606 Applied Deep Learning for Engineering - Course Syllabus
Lectures: Thursdays, 5:40 PM - 8:40 PM, Hill Center 009
Instructor:
Office:
Phone:
Email:
Website:
Office Hours:
Nikolaos N. Vlassis, Assistant Professor
ENG A-200
848-445-5517
nick.vlassis@rutgers.edu
https://canvas.rutgers.edu
Thursdays 2:00-4:00 PM, A200 and over Zoom (same as lecture link) and Mondays 1:00 PM – 2:00 PM over Zoom, or by appointment
Recommended Textbooks:
-
Ian Goodfellow, Yoshua Bengio, Aaron Courville, Deep Learning, MIT Press, available online at https://www.deeplearningbook.org
-
W. Michael Lai, David Rubin, Erhard Krempl, Introduction to Continuum Mechanics, Elsevier, 2010
Course Outline:
Lectures + Lab Sessions + Assignments + Final Project
Method of Evaluation:
Lab Sessions 30%
Homework 40%
Final Project 30%
Lecture and Lab Format
Every session will include a lecture component followed by a hands-on lab session focused on the topic discussed. Students will be required to code during the lab sessions to reinforce the material covered in the lectures. Students are asked to bring a device with internet access, preferably a laptop, to participate in the programming exercises. The lab sessions will use Google Colab, which runs in the cloud, so there is no need for a high-performance computer. All assignments and final projects will be completed using Python and PyTorch.
Lectures and lab sessions will be streamed live on Zoom, and recordings will be available online after the class ends. While attendance is not mandatory, students in the asynchronous section are welcome and encouraged to join the live sessions to benefit from real-time assistance during the lab session walkthroughs. This provides an opportunity to ask questions and receive live support while working on the coding exercises.
Topics covered:
Fundamentals
- Introduction to Deep Learning for Engineering Mechanics
- Basics of Python Programming and Scientific Computing
- Data Processing, Feature Engineering, and Visualization
- Introduction to PyTorch and Neural Network Implementation
- Deep Learning Pipeline: Data Collection, Training, Testing, and Integration of Physics
Neural Network Architectures
- Fundamentals of Neural Networks and Optimization
- Multilayer Perceptrons (MLPs)
- Convolutional Neural Networks (CNNs)
- Recurrent Neural Networks (RNNs)
- Data Structures: Tensors, Fields, Graphs, and Point Clouds
Applications in Engineering Mechanics
- Validation of Neural Network Models for Mechanics Applications
- Physics-informed constraints
- Constitutive Modeling and Material Characterization
- Physics-Based Solvers and Reduced-Order Modeling
- Optimization and Generative Design
Advanced Topics
- Reinforcement Learning for Control and Optimization
- Physics-Informed Neural Networks (PINNs)
- Geometric Learning and Graph Neural Networks
- Generative Models for Design and Mechanics Applications