Few-Shot Transfer Learning for Individualized Braking Intent Detection on Neuromorphic Hardware
This work addresses the need for personalized, energy-efficient brain-computer interfaces for braking intent detection, representing an incremental improvement over group-level models.
The paper tackled the problem of developing individual-specific braking intention predictive models from EEG data by using a few-shot transfer learning method on neuromorphic hardware, achieving at least 90% accuracy, true positive rate, and true negative rate with only three training epochs and a 97% power reduction compared to a CPU.
Objective: This work explores use of a few-shot transfer learning method to train and implement a convolutional spiking neural network (CSNN) on a BrainChip Akida AKD1000 neuromorphic system-on-chip for developing individual-level, instead of traditionally used group-level, models using electroencephalographic data. Main Results: Efficacy of the above methodology to develop individual-specific braking intention predictive models by rapidly adapting the group-level model in as few as three training epochs while achieving at least 90% accuracy, true positive rate and true negative rate is presented. Further, results show the energy-efficiency of the neuromorphic hardware through a power reduction of over 97% with only a $1.3* increase in latency when using the Akida AKD1000 processor for network inference compared to an Intel Xeon central processing unit. Similar results were obtained in a subsequent ablation study using a subset of five out of 19 channels.