Bio-Inspired Techniques in a Fully Digital Approach for Lifelong Learning

A survey and perspective on neuromorphic continual learning systems

With the advent of low-power neuromorphic computing systems, new possibilities have emerged for deployment in various sectors, like healthcare and transport, that require intelligent autonomous applications. These applications require reliable low-power solutions for sequentially adapting to new relevant data without loss of learning. Neuromorphic systems are inherently inspired by biological neural networks that have the potential to offer an efficient solution toward the feat of continual learning. With increasing attention in this area, we present a first comprehensive review of state-of-the-art neuromorphic continual learning (NCL) paradigms. The significance of our study is multi-fold. We summarize the recent progress and propose a plausible roadmap for developing end-to-end NCL systems. We also attempt to identify the gap between research and the real-world deployment of NCL systems in multiple applications. We do so by assessing the recent contributions in neuromorphic continual learning at multiple levels-applications, algorithms, architectures, and hardware. We discuss the relevance of NCL systems and draw out application-specific requisites. We analyze the biological underpinnings that are used for acquiring high-level performance. At the hardware level, we assess the ability of the current neuromorphic platforms and emerging nano-device-based architectures to support these algorithms in the presence of several constraints. Further, we propose refinements to continual learning metrics for applying them to NCL systems. Finally, the review identifies gaps and possible solutions that are not yet focused upon for deploying application-specific NCL systems in real-life scenarios.

A Brain-Inspired Homeostatic Neuron Based on Phase-Change Memories for Efficient Neuromorphic Computing

One of the main goals of neuromorphic computing is the implementation and design of systems capable of dynamic evolution with respect to their own experience. In biology, synaptic scaling is the homeostatic mechanism which controls the frequency of neural spikes within stable boundaries for improved learning activity. To introduce such control mechanism in a hardware spiking neural network (SNN), we present here a novel artificial neuron based on phase change memory (PCM) devices capable of internal regulation via homeostatic and plastic phenomena. We experimentally show that this mechanism increases the robustness of the system thus optimizing the multi-pattern learning under spike-timing-dependent plasticity (STDP). It also improves the continual learning capability of hybrid supervised-unsupervised convolutional neural networks (CNNs), in terms of both resilience and accuracy. Furthermore, the use of neurons capable of self-regulating their fire responsivity as a function of the PCM internal state enables the design of dynamic networks. In this scenario, we propose to use the PCM-based neurons to design bio-inspired recurrent networks for autonomous decision making in navigation tasks. The agent relies on neuronal spike-frequency adaptation (SFA) to explore the environment via penalties and rewards. Finally, we show that the conductance drift of the PCM devices, contrarily to the applications in neural network accelerators, can improve the overall energy efficiency of neuromorphic computing by implementing bio-plausible active forgetting.

Effect of repetition on the behavioral and neuronal responses to ambiguous Necker cube images

A repeated presentation of an item facilitates its subsequent detection or identification, a phenomenon of priming. Priming may involve different types of memory and attention and affects neural activity in various brain regions. Here we instructed participants to report on the orientation of repeatedly presented Necker cubes with high (HA) and low (LA) ambiguity. Manipulating the contrast of internal edges, we varied the ambiguity and orientation of the cube. We tested how both the repeated orientation (referred to as a stimulus factor) and the repeated ambiguity (referred to as a top-down factor) modulated neuronal and behavioral response. On the behavioral level, we observed higher speed and correctness of the response to the HA stimulus following the HA stimulus and a faster response to the right-oriented LA stimulus following the right-oriented stimulus. On the neuronal level, the prestimulus theta-band power grew for the repeated HA stimulus, indicating activation of the neural networks related to attention and uncertainty processing. The repeated HA stimulus enhanced hippocampal activation after stimulus onset. The right-oriented LA stimulus following the right-oriented stimulus enhanced activity in the precuneus and the left frontal gyri before the behavioral response. During the repeated HA stimulus processing, enhanced hippocampal activation may evidence retrieving information to disambiguate the stimulus and define its orientation. Increased activation of the precuneus and the left prefrontal cortex before responding to the right-oriented LA stimulus following the right-oriented stimulus may indicate a match between their orientations. Finally, we observed increased hippocampal activation after responding to the stimuli, reflecting the encoding stimulus features in memory. In line with the large body of works relating the hippocampal activity with episodic memory, we suppose that this type of memory may subserve the priming effect during the repeated presentation of ambiguous images.