“…The problem of inferring or searching for model parameters that match model behaviors to experimental constraints constitutes an inverse problem (Tarantola, 2016 ), for which analytical solutions rarely exist for complex dynamical systems, i.e., most mathematical models in neuroscience. Historically, such parameter searches were done by hand tuning, but the advent of increasingly powerful computing resources has brought automated search algorithms that can find suitable parameters (Bhalla and Bower, 1993 ; Vanier and Bower, 1999 ; Achard and De Schutter, 2006 ; Druckmann et al, 2007 ; Gurkiewicz and Korngreen, 2007 ; Van Geit et al, 2007 , 2008 ; Huys and Paninski, 2009 ; Taylor et al, 2009 ; Hay et al, 2011 ; Bahl et al, 2012 ; Svensson et al, 2012 ; Friedrich et al, 2014 ; Pozzorini et al, 2015 ; Stefanou et al, 2016 ). While many varieties of search algorithms have been described and explored in the literature (Vanier and Bower, 1999 ; Van Geit et al, 2008 ; Svensson et al, 2012 ), stochastic optimisation approaches, such as simulated annealing and evolutionary algorithms, have been shown to be particularly effective strategies for such parameter searches (Vanier and Bower, 1999 ; Druckmann et al, 2007 ; Gurkiewicz and Korngreen, 2007 ; Svensson et al, 2012 ).…”