Syllabus

Photo credit: Mateusz Zwierzycki

Description

This course will be an opportunity to experiment with various machine learning tools and algorithms available in Owl. It will start with a lecture explaining the current and the past problems of AI in design. The presentation will introduce the notions of supervised, unsupervised and reinforcement learning. Using Owl, the participants will then learn the Grasshopper-based workflow for neural networks (supervised learning). Introduction will end with an overview of Owl components used for data pre- and post-processing.

After the introduction, the focus will be put on the group projects and the ways to utilise ML to aid the designs either in sensing/responding, controlling and predicting. Neural networks will be used for both regression and classification problems. The students will learn how to understand and control the learning process through hyperparameters. 

Over the course of the semester this knowledge shall be extended to application of ANN in the students’ projects.

Learning Objectives

During this course the students will be challenged to incorporate artificial neural network models either for sensing/response, control or prediction. 

To be able to work efficiently with the NN models, the students will have to: 

  • understand the meaning of the training parameters 
  • understand the consequences of NN architecture changes 
  • learn how to prepare and post-process the data for ML

Faculty


Faculty Assistants


Projects from this course

Machine Learning to predict no. of seating spaces

Aim: To predict the number of seating spaces based on various types of seating layout, number of corridors and dimensions of generative enclosed rectangular spaces such as an auditorium. Objectives: Dataset Design: To create Dataset for the required problem, a synthetic dataset is designed with the help of various bylaws supporting the problem. Data Analysis: … Read more

Timber outlook

The project’s objective was to create a machine-learning model capable of classifying repurposed timber components within an assembly process as either structural or non-structural, using factors such as defect quantity, age, and exposure to weather conditions as input. Dataset Generation and Analysis The dataset was produced using Roboflow by utilizing scans of the timber elements … Read more

Predicting Ceramic Underglaze Colors

Our aim is to develop a machine learning model that accurately predicts the color outcome of ceramic underglazes based on their ingredient compositions and firing conditions. In the world of ceramic art, the process of underglazing involves applying colors to pottery which are then sealed under a transparent glaze before firing. However, predicting the final … Read more

Classification using Machine Learning

Our project involves the fabrication of a curved surface using cork panels that have been discretized into unique four-sided shapes. Our previous approach involved cutting each panel individually, but we aim to streamline the process by using machine learning to classify the panels into five distinct groups. We will then design and create a mold … Read more