In this seminar students will be introduced to conceptual perspectives around machine learning, dataset design and feature encoding.


Syllabus

Data Encoding Seminar 

MaCAD Digital Tools for DATA ENCODING & ML SEMINAR

Machine learning algorithms offer an alternative modelling paradigm for complex problems. Differently from analytical models where processes are computed using iterative time-based heavy calculations and pre-assigned property parameters, machine learning models learn by example through searching for an approximated relationship between the inputs and the targets. Once trained, they can be easily deployed within a design workflow to offer predictions. While these models offer advantages of accuracy and speed for iterative design, they are notoriously data-hungry. Many open source datasets are available online for training of state-of-the-art models, however these datasets are not relevant for architectural applications.

Dataset design is the enabler to using Machine Learning algorithms in different fields. While an existing ecology of algorithms provides architectures suitable to specific types of problems be classification or regression, the models must be trained on datasets that are specific to the problem they are expected to solve. They can be generated both computationally through heuristic algorithms, or by digitizing physical samples. Dataset design requires several criteria to be followed: Increased sample diversity, bias avoidance and ambiguity avoidance. These criteria ensure that the problem is well represented, with equal distribution and without confusion.


Source: Hesham Shawky & Cami Quinteros –  Imaginary Vessels. MaCAD 2021

In this seminar, you will be introduced to conceptual perspectives around machine learning, dataset design and feature encoding. Through direct hands-on experience, you will explore the usage of machine learning to predict geometry. You will be tasked to develop a computational model to produce a generative dataset. Through the study of the parameter space and feature distribution and representation, you will propose a dataset encoding method, and evaluate it through the training of shallow models as well as artificial neural networks, and represent the predicted outcome in the design space. 

Learning Objectives 

At course completion the student will:

  •       Introduced to fundamental machine learning concepts and workflows.
  •       Understand notions of parameter space, data encoding and feature selection
  •       Acquire competences in developing generative datasets for architectural application
  •       Acquire competences in feature engineering such as dimensionality reduction, data analysis
  •       Acquire a hands-on experience in using state of the art machine learning libraries
  •       Develop appropriate representational methods and tools to showcase your findings
  •       Collaborate effectively within an online group-working context

Faculty


Faculty Assistants


Projects from this course

Skylight Performance

This project studies skylight performance in four given cities: Barcelona, Malina, Melbourne, and Vancouver collectively, by employing of Data-Encoding machine learning methods. ——————————————————————————————————————————————————————————————————————————- ——————————————————————————————————————————————————————————————————————————- ——————————————————————————————————————————————————————————————————————————- ——————————————————————————————————————————————————————————————————————————- ——————————————————————————————————————————————————————————————————————————- ——————————————————————————————————————————————————————————————————————————- ——————————————————————————————————————————————————————————————————————————- ——————————————————————————————————————————————————————————————————————————- ——————————————————————————————————————————————————————————————————————————- ——————————————————————————————————————————————————————————————————————————- Summary of Execution ‘Data_Encoding’ machine learning method offers renovative means for a variety of applications in architectural design. In order to have a … Read more

ESTIMATE ESTATE – Discovering the Perfect Locale

With the recent upsurge in tourism worldwide, the Airbnb accommodation sector has become a thriving industry. As a result, the demand for Airbnb accommodations is poised for further growth, attracting an increasing number of individuals interested in entering this lucrative market. ●How can a new entrepreneur make a decision on the ideal location to open … Read more

Floor2View

What is an isovist? Isovist analysis helps architects and designers understand how people perceive and navigate through spaces, enabling themto optimize layouts, enhance wayfinding, and create more engaging and functional environments. This project has only work over the isovist area calculation, in which the final isofield image represents each of the points of view visible … Read more

TERRADOME

“Unlocking New Horizons in Immersive Acoustic Environments!” Traditional methods of predicting sound acoustics in domes require extensive manual calculations and simulations, leading to time-consuming and resource-intensive processes. The idea is to develop a machine learning model that could calculate the complex acoustic properties of dome structures. This model can help us to predict the sound … Read more

Corbel Arches

We decided to study how machine learning can help us to stack stones and build corbel arches. Corbel arches are a type of arch structure in which the arch is formed by successively projecting stones or bricks from each side until they meet at the center. The arch relies on the structural integrity of the … Read more

APARTMENT PRICE PREDICTION

Based on some features of an urban neighborhood can a Machine Learning model predict the price of a property? In response to the mandate: develop a project integrating urban data with geometric dimensions, our team elected to ascertain unit prices via a selection of predetermined metrics. Initially, our targeted location was within the central region … Read more

Mississippi 2102

Meanders on the Lower Mississippi River In the battle between architecture and nature, nature tends to win. Therefore, it is important that we understand how rivers move over time so that we can know the best and the worst places to build. Our project studies the Lower Mississippi River from Cairo, Illinois to New Orleans, … Read more

MASHRABIYA

ABSTRACT Mashraraya is an essential element of Islamic architecture, and its significance surpasses the aesthetics and symbolism. It serves as more than just a decorative feature, protecting the interior spaces from harsh weather by providing shading and encouraging natural ventilation. It also offers privacy and can be used as a structural element. The project’s objective … Read more

Perfect Seat

Find the perfect seat in a stadium! Problem Statement How many times have you been to a stadium, sports complex, concert or related events where your seat look much less appealing  than in the sales chart? This raises a very important question, what are the metrics used to define the quality of a seat in … Read more

What makes a park popular?

This research revolves around a urban dilemma. Everyone can think of a park in their city that is very crowded and another one that is always desert. why does this happen? Is it related to the design of the park? Our objective is to Look for a relation into Park Features/Popularity. If there is a … Read more

Adappt

Challenge to solve The potential for a building to adapt to various uses is primarily determined during the initial stages of the design process. This feature is rarely considered by architects yet has a significant impact of a building’s sustainability. The intention is to provide an early-stage tool that architects can use to test adaptability … Read more