Course on computational design, non-linear optimization, and dynamics of soft systems at UIUC.

Last update: Jul 21, 2022

Overview

Computational Design and Dynamics of Soft Systems ·

This is a repository that contains the source code for generating the lecture notes, handouts, exercises for the computational lab-sessions of the course offered at UIUC.

Description

This course provides a hands-on introduction to modern modeling and simulations techniques for heterogeneous structures made of assemblies of soft, elastic slender elements. Such systems are ubiquitous in nature, from animal musculoskeletal architectures to ‘birds-nest’ composite materials. They are also becoming increasingly relevant in robotics. Students will implement in python their own Cosserat rods-based solver. The developed solver will be then coupled with evolutionary optimization techniques for design, and reinforcement learning for control.

Prerequisities

None.

Content

Introduction to modeling and simulation for inverse design
Basics of evolutionary strategies
Covariance Matrix Adaptation – Evolution Strategy (CMA-ES)
Basic concepts of Reinforcement Learning
Soft robotic modeling with Cosserat rods
Space and time discretization
Application to snake slithering
Complex creatures modeling
Examples of potential experimental applications

Organization

The course is organized in three modules listed below.

Python for engineers
- Crash course in Python for engineers
  - Scientific computing via Python
Non-linear stochastic optimization
- Implementing CMA-es for nonlinear stochastic optimization
  - Adopting CMA-es to tackle real-life inverse-design problems
Modeling of soft systems
Visualizing soft-system dynamics

Setup

To get started with the course, please consult this folder.

Course on computational design, non-linear optimization, and dynamics of soft systems at UIUC.

Related tags

Overview

Computational Design and Dynamics of Soft Systems ·

Description

Prerequisities

Content

Organization

Setup

Owner

Tejaswin Parthasarathy

Check out the StyleGAN repo and place it in the same directory hierarchy as the present repo

UI2I via StyleGAN2 - Unsupervised image-to-image translation method via pre-trained StyleGAN2 network

PatrickStar enables Larger, Faster, Greener Pretrained Models for NLP. Democratize AI for everyone.

Disease Informed Neural Networks (DINNs) — neural networks capable of learning how diseases spread, forecasting their progression, and finding their unique parameters (e.g. death rate).

Official Implementation of 'UPDeT: Universal Multi-agent Reinforcement Learning via Policy Decoupling with Transformers' ICLR 2021(spotlight)

COVINS -- A Framework for Collaborative Visual-Inertial SLAM and Multi-Agent 3D Mapping

Implementation of Continuous Sparsification, a method for pruning and ticket search in deep networks

Official implementation of "GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators" (NeurIPS 2020)

tsai is an open-source deep learning package built on top of Pytorch & fastai focused on state-of-the-art techniques for time series classification, regression and forecasting.

Automatic tool focused on deriving metallicities of open clusters

Relative Uncertainty Learning for Facial Expression Recognition

Video Frame Interpolation with Transformer (CVPR2022)

StarGAN v2 - Official PyTorch Implementation (CVPR 2020)

Differentiable Wavetable Synthesis

An image processing project uses Viola-jones technique to detect faces and then use SIFT algorithm for recognition.

Mercer Gaussian Process (MGP) and Fourier Gaussian Process (FGP) Regression

TART - A PyTorch implementation for Transition Matrix Representation of Trees with Transposed Convolutions

Awesome Weak-Shot Learning

ScaleNet: A Shallow Architecture for Scale Estimation

DeeBERT: Dynamic Early Exiting for Accelerating BERT Inference