Efficient Reward-Based Structural Plasticity on a SpiNNaker 2 Prototype

Abstract

Advances in neuroscience uncover the mechanisms employed by the brain to efficiently solve complex learning tasks with very limited resources. However, the efficiency is often lost when one tries to port these findings to a silicon substrate, since brain-inspired algorithms often make extensive use of complex functions such as random number generators, that are expensive to compute on standard general purpose hardware. The prototype chip of the 2nd generation SpiNNaker system is designed to overcome this problem. Low-power ARM processors equipped with a random number generator and an exponential function accelerator enable the efficient execution of brain-inspired algorithms. We implement the recently introduced reward-based synaptic sampling model that employs structural plasticity to learn a function or task. The numerical simulation of the model requires to update the synapse variables in each time step including an explorative random term. To the best of our knowledge, this is the most complex synapse model implemented so far on the SpiNNaker system. By making efficient use of the hardware accelerators and numerical optimizations the computation time of one plasticity update is reduced by a factor of 2. This, combined with fitting the model into to the local SRAM, leads to 62% energy reduction compared to the case without accelerators and the use of external DRAM. The model implementation is integrated into the SpiNNaker software framework allowing for scalability onto larger systems. The hardware-software system presented in this work paves the way for power-efficient mobile and biomedical applications with biologically plausible brain-inspired algorithms.

On-Chip Error-Triggered Learning of Multi-Layer Memristive Spiking Neural Networks

“…In this work, we target the general multi-layer learning problem by taking into account the neural dynamics and multiple layers. Currently, Intel Loihi research chip, Spinnaker 1 and 2, and the Brainscales-2 have the ability to implement a vast variety of learning rules [1], [29], [30]. Spinnaker and Loihi are both research tools that provide a flexible programmable substrate that can implement a vast set of learning algorithms.…”

Plasticity and Adaptation in Neuromorphic Biohybrid Systems

“…Beyond such methods of LTP that focus on the manipulation of weights in a predefined network topology, the optimization of synaptic resources through structural plasticity can be achieved in neuromorphic circuits as illustrated in George et al, 2017 and Yan et al. (2019) .…”

“…Beyond such methods of LTP that focus on the manipulation of weights in a predefined network topology, the optimization of synaptic resources through structural plasticity can be achieved in neuromorphic circuits as illustrated in George et al, 2017 andYan et al (2019). In George et al, 2017 the authors expand an STDP-based learning rule for the simulation of synaptogenesis (here the routing of a new pre-synaptic partner to the synapse circuit connected to the post-synaptic neuron) and for synapse degeneration (the removal of the pre-synaptic partner connection).…”

Embodied tactile perception and learning

“…Recently, SpiNNaker-related research has attracted wide attention from several countries, including the United Kingdom, France, Germany, United States, and Japan. On a SpiNNaker system, a structural plasticity learning function is used to implement a reward-based synapse sampling algorithm that provides an effective tool for brain-inspired algorithms [138]. Serrano-Gotarredona et al [139] optimized the storage design in the SpiNNaker system based on the "weight sharing" feature of a CNN by implementing a five-layer CNN for symbol recognition.…”