T. Sibindi

Lightweight Reinforcement Algorithms for autonomous, scalable intra-cortical Brain Machine Interfaces

This paper introduces Banditron, a lightweight RL algorithm for iBMIs, showing significant power savings and performance improvements over state-of-the-art methods.

Towards Autonomous Intra-cortical Brain Machine Interfaces: Applying Bandit Algorithms for Online Reinforcement Learning

This paper applies Banditron, an online reinforcement learning algorithm, to improve decoding performance in intra-cortical brain machine interfaces using non-human primate data.

Sparse Ensemble Machine Learning to improve robustness of long-term decoding in iBMIs

This paper introduces a sparse ensemble machine learning method to improve decoding robustness in iBMIs by reducing sensitivity to recording channel variations, achieving up to 24% accuracy improvement in non-human primates.

Towards Intelligent Intra-cortical BMI (i2BMI):Low-power Neuromorphic Decoders that outperform Kalman Filters

This paper introduces a low-power neuromorphic decoder, SELMA, which outperforms Kalman filters in continuous state decoding for intra-cortical BMI systems.