Image source: Course faculty
Evolutionary Computation is a branch of computation that is quite unique. For one, it is not specific to any problem or task. It is a framework for solving generic problems. This offers interesting capacities for the design process where we are usually the ones iterating over a design decision in order to evaluate its effectiveness. If we could abstract the forces which guide a design decision, then we could potentially utilize evolutionary computation, and specifically Genetic algorithms, to assist us in finding optimal solutions given a number of design criteria. The Designing Associativity seminar will take a look at how we set up a design space that is analytical in nature. We will also look at how to tailor this design space to evolve optimized design solutions. Students would be given intense workshop on learning of these software's and then studio time would be given to experiment the digital scripting basis of their designs. Task is to utilize Evolutionary Computation in designs. Define a design task to abstract, define a solution space, and deploy Galapagos to search the solution space for the most optimal solutions. You are to define clearly the parameters and how you are trying to maximize, minimize, or reach a certain fitness. We will also see how to connect Galapagos with Heliotrope, so this is also an option for those looking to utilize environmental forces as a driver for design.Skills to be taught (i) Advanced design approach (ii) Associative design (iii) Feed Loop. Goutham Santhanam (masters in advanced architecture, IAAC, Barcelona)- will be the academic associate for the course and will accompany to help in digital scripting. Expenses for students: Material cost: Rs. 500/-.