Machine learning in grasshopper. Warsaw

90% of the data in the world was produced just over the last two years. Each day we produce around 2.5 quintillion bytes of data via social networks, search engines, sensors and more. Machine learning algorithms allow the in depth analysys of the data in order organize its structure, understand the data and to predict processes in the future.

Construction and building big data already exists in all the plans and records of anything that was ever built. During our design process, as computational designers, we deal everyday with a large data sets. Each optimization algorithm, each desicion we take, each solution we find is a set of data that can be stored and used in the future. We do automize the design process, but we also know that there are decisions that computer cannot make on its own and which are unique to the architect and based on his preferences.

What if we can teach algorithm, how to mimic our own decision making process? What if an algorithm can predict the impact of our decisions on the construction process and the building life cycle? And finally, what if the computer become our partner in the design process, not only following the instructions we write, but creating its own design solutions?

The 16 hours long course will focus on the theoretical background of machine learning algorithms, practical examples of selected algorithms in the Grasshopper environment and its possible application in the design process.

What you will learn:

- Basics of machine learning concepts (Supervised and Unsupervised Learning)
- Data mining
- Data preparation
- Selected of machine learning algorithms:
     Linear Regression
     K-means Clustering
     T-SNE
     Artificial Neural Networks
- Application of Machine Learning in the architectural design process
- Application of Machine Learning in the optimization process

We will work on the various design scales: urban, building, ornamen detail and simulation.

Tutor:
Kacper Radziszewski

REQUIREMENTS

- Basic knowledge of Grasshopper is essential

- Selected libraries installed (instructions will be provided before the event)

- Laptop

COSTS:
800 PLN (200 eur) students

1020 PLN (270 eur) professionals