Recommended but not required: Linear Algebra, Analysis, Differential Equations, Neurodynamics, TensorFlow
In the first half of this study project, we became familiar with the concept of computation at the edge of chaos (Langton 1990). We got acquainted with its applications within the reservoir computing arena
(Bertschinger et. al, 2004), where it was shown that, for reservoirs with binary threshold units, performance peaked when the reservoir weights were scaled to just below a critical value, after which state convergence was lost. We also examined the criticism that for analog echo-state networks, computation at the edge of chaos offers no general benefit (Yildiz et. al, 2012).
In the second half of the project, we will further investigate the concept of computation at the edge of chaos from different angles. We will divide our time between deep feedforward architectures, reservoir computing and models from theoretical neuroscience. At the end of the project, we will try to write a comprehensive review of the research into computation at the edge of chaos."