Agent based systems coupled with machine learning algorithms
Current on-going research focusing on unsupervised learning algorithms in order to classify and cluster behavioral inputs and integrate them within the design process, as a form-finding phase through a non-linear morph of behavior parameters.
AGENT BASED SYSTEMS COUPLED WITH MACHINE LEARNING ALGORITHMS
Self-organizing maps (SOM) are used here, (SOM represents a type of unsupervised artificial neural network), to reduce the high dimensionality of the data whilst also retain the high-dimensional non-linear associations. As SOMs can create associations between inputs, the map can suggest an overview of possibilities within the given parameter space, without the need of manually tweaking the system’s parameters by the designer.
The selected 8 models are used as inputs for the SOM. The resulting map creates an interpolation between inputs and creates a location rule, placing similar models closer on the map while placing further apart dissimilar models.
AGENT BASED SYSTEM TRACE FOLLOW BEHAVIOR PARAMETERS
COESHION VIEW RANGE
SEPARATION VIEW RANGE
FIELD MIN INTENSITY
FIELD MAX INTENSITY
OVERALL SYSTEM LENGTH
BEHAVIOR PARAMETERS INFLUENCE DISTRIBUTION
K-mean clustering is used here to analyze parameter distributions on the self-organizing map (SOM), and help cluster and classify insightful behavioral patterns, and quantify the influence each has on the outcome.
DEEP LEARNING Research augmented creativity design space explorer 3d domain translations Computer vision generative adversarial networks neural networks MACHINE LEARNING dimensionality reduction ADVANCED Computational Design genetic algorithms generative algorithms AGENT BASED SYSTEMS multi-agent systems structural encoded agents