A reconfigurable on-line learning spiking neuromorphic processor comprising 256 neurons and 128K synapses

Abstract

Implementing compact, low-power artificial neural processing systems with real-time on-line learning abilities is still an open challenge. In this paper we present a full-custom mixed-signal VLSI device with neuromorphic learning circuits that emulate the biophysics of real spiking neurons and dynamic synapses for exploring the properties of computational neuroscience models and for building brain-inspired computing systems. The proposed architecture allows the on-chip configuration of a wide range of network connectivities, including recurrent and deep networks, with short-term and long-term plasticity. The device comprises 128 K analog synapse and 256 neuron circuits with biologically plausible dynamics and bi-stable spike-based plasticity mechanisms that endow it with on-line learning abilities. In addition to the analog circuits, the device comprises also asynchronous digital logic circuits for setting different synapse and neuron properties as well as different network configurations. This prototype device, fabricated using a 180 nm 1P6M CMOS process, occupies an area of 51.4 mm2, and consumes approximately 4 mW for typical experiments, for example involving attractor networks. Here we describe the details of the overall architecture and of the individual circuits and present experimental results that showcase its potential. By supporting a wide range of cortical-like computational modules comprising plasticity mechanisms, this device will enable the realization of intelligent autonomous systems with on-line learning capabilities.

Predicting voluntary movements from motor cortical activity with neuromorphic hardware

“…SpiNNaker [24], or the low-power analog ROLLS neuromorphic processor [26]. These platforms would be more straightforward to use in a closed-loop scenario, because networks on these chips will operate at exactly the same time scale as the brain, and recorded spikes can be injected into the network in real-time.…”

“…Several platforms for neuromorphic computing have emerged, starting with the pioneering work of Carver Mead [23], up to rather recent developments like e.g. SpiNNaker [24], NeuroGrid [25], ROLLS [26], IBM's TrueNorth [27], or the systems developed at the University of Heidelberg [28], [29] (see [30] for a review). Aside from testing principles of neural computation [28], [31], [32], initial applications of neuromorphic hardware have been demonstrated for generic pattern recognition [27], [33].…”

Neuromorphic implementations of neurobiological learning algorithms for spiking neural networks

“…One of the most recent chip designs presented in this review was developed by Qiao et al (2015). The Reconfigurable On-line Learning Spiking (ROLLS) neuromorphic processor contains 256 analog silicon neurons and a total of 131 072 synapses.…”

“…Like its predecessors, the ROLLS chip is able to operate in biological real-time. The power consumption in typical scenarios is 4 mW (Qiao et al, 2015). Special focus has been put on the implementation of different synaptic plasticity mechanisms, which will be discussed in Section 4.3.…”

“…The optimal connectivity pattern for a given task can be inferred by evolutionary processes and genetic programming, which are both beyond the scope of this paper. However, it should be mentioned that the ROLLS neuromorphic processor can change its connectivity at runtime (Qiao et al, 2015), which is an important prerequisite for implementing an evolutionary algorithm in an autonomous robot. In this section, we provide a brief and non-exhaustive overview of learning techniques for spiking neural networks.…”