Hands-On Parameter Search for Neural Simulations by a MIDI-Controller

Hubert Eichner,Alexander Borst

doi:10.1371/journal.pone.0027013

Abstract

Computational neuroscientists frequently encounter the challenge of parameter fitting – exploring a usually high dimensional variable space to find a parameter set that reproduces an experimental data set. One common approach is using automated search algorithms such as gradient descent or genetic algorithms. However, these approaches suffer several shortcomings related to their lack of understanding the underlying question, such as defining a suitable error function or getting stuck in local minima. Another widespread approach is manual parameter fitting using a keyboard or a mouse, evaluating different parameter sets following the users intuition. However, this process is often cumbersome and time-intensive. Here, we present a new method for manual parameter fitting. A MIDI controller provides input to the simulation software, where model parameters are then tuned according to the knob and slider positions on the device. The model is immediately updated on every parameter change, continuously plotting the latest results. Given reasonably short simulation times of less than one second, we find this method to be highly efficient in quickly determining good parameter sets. Our approach bears a close resemblance to tuning the sound of an analog synthesizer, giving the user a very good intuition of the problem at hand, such as immediate feedback if and how results are affected by specific parameter changes. In addition to be used in research, our approach should be an ideal teaching tool, allowing students to interactively explore complex models such as Hodgkin-Huxley or dynamical systems.

Highlights

A frequent challenge in computational neuroscience is to come up with models that properly replicate some quantitative or qualitative characteristics of measured data
While successfully applied in a large range of projects, these methods suffer from several shortcomings, such as requiring an exactly defined measure of quality for a given parameter set, slow convergence or ending up in local minima
Standard error functions like the root mean squared difference between the actual and the target output may fail to arrive at suitable parameter sets, and significant effort is spent on defining and refining a proper error function

Summary

Introduction

A frequent challenge in computational neuroscience is to come up with models that properly replicate some quantitative or qualitative characteristics of measured data. A profound problem, becomes apparent when the user desires a good qualitative, not quantitative fit, for instance because the model is intentionally simplified and unable to exactly reproduce the results, or because the desired behavior of the model is hard to formulate in mathematical terms. In such cases, standard error functions like the root mean squared difference between the actual and the target output may fail to arrive at suitable parameter sets, and significant effort is spent on defining and refining a proper error function. A further problem is optimizing multiple objectives at the same time [5] when the tradeoff between the different objectives depends on the user’s intuition

Methods

Results

Conclusion