1 code implementation • 4 May 2020 • Fabio Colella, Pedram Daee, Jussi Jokinen, Antti Oulasvirta, Samuel Kaski
To verify this hypothesis, that humans steer and are able to improve performance by steering, we designed a function optimization task where a human and an optimization algorithm collaborate to find the maximum of a 1-dimensional function.
1 code implementation • 1 Nov 2019 • Tomi Peltola, Jussi Jokinen, Samuel Kaski
The strengths of the probabilistic formulation, in addition to providing a bounded rational account of the learning of the heuristic, include natural extensibility with additional cognitively plausible constraints and prior information, and the possibility to embed the heuristic as a subpart of a larger probabilistic model.