Black-box and surrogate optimization for tuning spiking neural models of striatum plasticity

Abstract

The basal ganglia (BG) is a brain structure that has long been proposed to play an essential role in action selection, and theoretical models of spiking neurons have tried to explain how the BG solves this problem. A recently proposed functional and biologically inspired network model of the striatum (an important nucleus of the BG) is based on spike-timing-dependent eligibility (STDE) and captured important experimental features of this nucleus. The model can recognize complex input patterns and consistently choose rewarded actions to respond to such sensory inputs. However, model tuning is challenging due to two main reasons. The first is the expert knowledge required, resulting in tedious and potentially biased trial-and-error procedures. The second is the computational cost of assessing model configurations (approximately 1.78 h per evaluation). This study addresses the model tuning problem through numerical optimization. Considering the cost of assessing solutions, the selected methods stand out due to their low requirements for solution evaluations and compatibility with high-performance computing. They are the SurrogateOpt solver of Matlab and the RBFOpt library, both based on radial basis function approximations, and DIRECT-GL, an enhanced version of the widespread black-box optimizer DIRECT. Besides, a parallel random search serves as a baseline reference of the outcome of opting for sophisticated methods. SurrogateOpt turns out to be the best option for tuning this kind of model. It outperforms, on average, the quality of the configuration found by an expert and works significantly faster and autonomously. RBFOpt and the random search share the second position, but their average results are below the option found by hand. Finally, DIRECT-GL follows this line becoming the worst-performing method.

Automatic placement of cameras in a particle accelerator using a 3D engine and computational optimization

“…To support the previous statement, we launched a random search [16,40] over the same problem. This method generates and evaluates random points in the search space while keeping the best solution found so far.…”

“…Besides, it can be easily coupled with high-performance computing, which allows increasing the number of solutions explored per unit of time. Hence, it may obtain acceptable results depending on the problem and the expectations to meet, as seen in [16,40].…”

“…As introduced, this context is suitable for addressing the problem through blackbox or derivative-free optimization methods [15,16,37]. They evaluate candidate solutions without further requirements from the objective function and obtain acceptable results with a reasonable effort.…”

“…Nevertheless, converting a maximization problem into a minimization one is straightforward. Specifically, maximizing an objective function f is equivalent to minimizing −f [16].…”

“…As mentioned, problem-independent heuristics (meta-heuristics) are widespread for optimization. Many exist [12,14], and their generic design makes them what is known as black-box optimization methods [11,15,16]. This name emphasizes how these algorithms only need to evaluate different candidate solutions, using the objective function as a black box without calculating derivatives.…”