A Digital Neuromorphic Architecture Efficiently Facilitating Complex Synaptic Response Functions Applied to Liquid State Machines
This work addresses energy and time inefficiencies in neural network simulations for AI and neuroscience applications, representing an incremental advancement in neuromorphic computing.
The paper tackles the computational bottleneck of simulating neural networks by introducing a digital neuromorphic architecture, the Spiking Temporal Processing Unit (STPU), which efficiently models complex synaptic response functions without extra hardware, achieving improved efficiency in speech recognition tasks with liquid state machines.
Information in neural networks is represented as weighted connections, or synapses, between neurons. This poses a problem as the primary computational bottleneck for neural networks is the vector-matrix multiply when inputs are multiplied by the neural network weights. Conventional processing architectures are not well suited for simulating neural networks, often requiring large amounts of energy and time. Additionally, synapses in biological neural networks are not binary connections, but exhibit a nonlinear response function as neurotransmitters are emitted and diffuse between neurons. Inspired by neuroscience principles, we present a digital neuromorphic architecture, the Spiking Temporal Processing Unit (STPU), capable of modeling arbitrary complex synaptic response functions without requiring additional hardware components. We consider the paradigm of spiking neurons with temporally coded information as opposed to non-spiking rate coded neurons used in most neural networks. In this paradigm we examine liquid state machines applied to speech recognition and show how a liquid state machine with temporal dynamics maps onto the STPU-demonstrating the flexibility and efficiency of the STPU for instantiating neural algorithms.