Acceleration of the SPADE Method Using a Custom-Tailored FP-Growth Implementation

Abstract

The SPADE (spatio-temporal Spike PAttern Detection and Evaluation) method was developed to find reoccurring spatio-temporal patterns in neuronal spike activity (parallel spike trains). However, depending on the number of spike trains and the length of recording, this method can exhibit long runtimes. Based on a realistic benchmark data set, we identified that the combination of pattern mining (using the FP-Growth algorithm) and the result filtering account for 85–90% of the method's total runtime. Therefore, in this paper, we propose a customized FP-Growth implementation tailored to the requirements of SPADE, which significantly accelerates pattern mining and result filtering. Our version allows for parallel and distributed execution, and due to the improvements made, an execution on heterogeneous and low-power embedded devices is now also possible. The implementation has been evaluated using a traditional workstation based on an Intel Broadwell Xeon E5-1650 v4 as a baseline. Furthermore, the heterogeneous microserver platform RECS|Box has been used for evaluating the implementation on two HiSilicon Hi1616 (Kunpeng 916), an Intel Coffee Lake-ER Xeon E-2276ME, an Intel Broadwell Xeon D-D1577, and three NVIDIA Tegra devices (Jetson AGX Xavier, Jetson Xavier NX, and Jetson TX2). Depending on the platform, our implementation is between 27 and 200 times faster than the original implementation. At the same time, the energy consumption was reduced by up to two orders of magnitude.

Comparing Surrogates to Evaluate Precisely Timed Higher-Order Spike Correlations

“…When analyzing large-size experimental data, the FIM search for all possible patterns can result in obtaining millions, if not billions, of putative patterns ( Porrmann et al, 2021 ). To reduce computation time, which is particularly relevant for large datasets, we require a minimum occurrence count min occ of each pattern size to be further considered after the frequent mining step.…”

“…The SPADE method and all implementations of the surrogate techniques are included in the Elephant Python package http://python-elephant.org . Regarding the computational cost, several improvements have been made for the performance of both, SPADE and the surrogate implementations ( https://elephant.readthedocs.io/en/latest/release_notes.html ; Porrmann et al, 2021 ). Nonetheless, depending on the size of the dataset and the number of surrogates employed, large analyses can still take up to several hours on a computer cluster.…”

Model of multiple synfire chains explains cortical spatio-temporal spike patterns

“…The first is the detection of all putative patterns in the data at a certain temporal resolution, during which the putative patterns are stored along with the information on when and how often they occur. This is achieved by applying a Frequent Itemset Mining algorithm [52,53]. The second step is the statistical evaluation of the significance of the patterns detected in the first step, under the null-hypothesis of mutual independence of spike trains given their firing rate (co-)modulations [54].…”