Speech2Spikes: Efficient Audio Encoding Pipeline for Real-time Neuromorphic Systems

Stewart, Kenneth; Shea, Timothy M.; Pacik-Nelson, Noah; Gallo, Eric M; Danielescu, Andreea

doi:10.1145/3584954.3584995

Cited by 8 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lastly, while we designed our method with the requirements of edge hardware in mind, an implementation on such a device is missing. Unlike many existing methods that are trained offline and allow for inference on neuromorphic hardware (Tang et al, 2020;Taunyazov et al, 2020;Stewart et al, 2023), local learning rules theoretically extend their applicability to neuromorphic processors during the training phase. The local learning rule that alleviates the need for backpropagation renders CML compatible with on-chip training while introducing a minimal memory footprint since weight updates can be done in-place.…”

Section: Limitations and Future Workmentioning

confidence: 99%

Mapless mobile robot navigation at the edge using self-supervised cognitive map learners

Polykretis,

Danielescu

2024

Front. Robot. AI

View full text Add to dashboard Cite

Navigation of mobile agents in unknown, unmapped environments is a critical task for achieving general autonomy. Recent advancements in combining Reinforcement Learning with Deep Neural Networks have shown promising results in addressing this challenge. However, the inherent complexity of these approaches, characterized by multi-layer networks and intricate reward objectives, limits their autonomy, increases memory footprint, and complicates adaptation to energy-efficient edge hardware. To overcome these challenges, we propose a brain-inspired method that employs a shallow architecture trained by a local learning rule for self-supervised navigation in uncharted environments. Our approach achieves performance comparable to a state-of-the-art Deep Q Network (DQN) method with respect to goal-reaching accuracy and path length, with a similar (slightly lower) number of parameters, operations, and training iterations. Notably, our self-supervised approach combines novelty-based and random walks to alleviate the need for objective reward definition and enhance agent autonomy. At the same time, the shallow architecture and local learning rule do not call for error backpropagation, decreasing the memory overhead and enabling implementation on edge neuromorphic processors. These results contribute to the potential of embodied neuromorphic agents utilizing minimal resources while effectively handling variability.

show abstract

Section: Limitations and Future Workmentioning

confidence: 99%