Sfi Postdoc Working on Neural Network Models of Social Organizations

Complex Systems Society; Santa Fe (NM); On-site; USD 60,000 - 75,000; Full time

Job summary

The Santa Fe Institute is seeking a postdoc to help lead a project that uses information theory and numerical optimization to analyze social organizations, ranging from modern firms to military structures to chiefdoms and states.

The project will apply deep learning techniques to identify the “organization” of a social group as its communication network, with a group-selection perspective and an exogenous welfare function for the overall group.

Job description

Sfi Postdoc Working on Neural Network Models of Social Organizations

We are looking for a postdoc to play a key role in a project using modern information theory and numerical optimization techniques to analyze social organizations, ranging from modern firms to military organizations to complex chiefdoms to primary states.

Location: Santa Fe Institute,USA

Duration: 2 years

theory of the firm network coding theory social organizations information theory

The starting point for the project is to identify the "organization" of a social group as the communication network(s) within that group. To begin we will adopt a group-selection perspective, assuming that the social group's network structure is (approximately) optimal, given the information-processing limitations of the agents within the social group, and the exogenous welfare function of the overall group. We intend to leverage the computational power of "deep learning" (i.e., neural network training algorithms) to solve such problems. An expanded (though now slightly out of date) description of this project can be found athttps://arxiv.org/abs/1702.04449 .

This project, led by the Santa Fe Institute, is a collaboration amongarchaeologists, economists, andcomputer scientists. The idealstarting date is the spring of 2018,though that is flexible. The position would last for two years, with possible extensions. Anyone who is interested please send an email to David Wolpert,[e-mail], including a c.v. and a brief description of yourself. Later in the process you may be asked to arrange for some letters of recommendation to be sent as well.