Neuromorphic computing shows promise for advancing computing efficiency and capabilities of AI applications using brain-inspired principles. However, the neuromorphic research field currently lacks standardized benchmarks, making it difficult to accurately measure technological advancements, compare performance with conventional methods, and identify promising future research directions. Prior neuromorphic computing benchmark efforts have not seen widespread adoption due to a lack of inclusive, actionable, and iterative benchmark design and guidelines. To address these shortcomings, we present NeuroBench: a benchmark framework for neuromorphic computing algorithms and systems. NeuroBench is a collaboratively-designed effort from an open community of researchers across industry and academia, aiming to provide a representative structure for standardizing the evaluation of neuromorphic approaches. The NeuroBench framework introduces a common set of tools and systematic methodology for inclusive benchmark measurement, delivering an objective reference framework for quantifying neuromorphic approaches in both hardware-independent (algorithm track) and hardware-dependent (system track) settings. In this article, we outline tasks and guidelines for benchmarks across multiple application domains, and present initial performance baselines across neuromorphic and conventional approaches for both benchmark tracks. NeuroBench is intended to continually expand its benchmarks and features to foster and track the progress made by the research community.

Edge devices operating in dynamic environments critically need the ability to continually learn without catastrophic forgetting. The strict resource constraints in these devices pose a major challenge to achieve this, as continual learning entails memory and computational overhead. Crossbar architectures using memristor devices offer energy efficiency through compute-in-memory and hold promise to address this issue. However, memristors often exhibit low precision and high variability in conductance modulation, rendering them unsuitable for continual learning solutions that require precise modulation of weight magnitude for consolidation. Current approaches fall short to address this challenge directly and rely on auxiliary high-precision memory, leading to frequent memory access, high memory overhead, and energy dissipation. In this research, we propose probabilistic metaplasticity, which consolidates weights by modulating their update probability rather than magnitude. The proposed mechanism eliminates high-precision modification to weight magnitudes and, consequently, the need for auxiliary high-precision memory. We demonstrate the efficacy of the proposed mechanism by integrating probabilistic metaplasticity into a spiking network trained on an error threshold with low-precision memristor weights. Evaluations of continual learning benchmarks show that probabilistic metaplasticity achieves performance equivalent to state-of-the-art continual learning models with high-precision weights while consuming ~ 67% lower memory for additional parameters and up to ~ 60x lower energy during parameter updates compared to an auxiliary memory-based solution. The proposed model shows potential for energy-efficient continual learning with low-precision emerging devices.

Crossbar architectures utilizing memristor devices hold promise to address continual learning challenges in resource-constrained edge devices. However, these nanoscale devices often exhibit low precision and high variability in conductance modulation, rendering them unsuitable for continual learning solutions that consolidate weights through precise modulation. This issue can be circumvented by accumulating weight gradients in auxiliary high-precision memory and updating memristor weights when gradients are equivalent to memristor weight resolution. However, it leads to frequent memory access, high memory overhead, and energy dissipation. In this research, we propose probabilistic metaplasticity, which consolidates weights by modulating their update probability rather than magnitude. The proposed mechanism eliminates high-precision modification to weight magnitude and consequently, high-precision memory for gradient accumulation. We demonstrate the efficacy of the proposed mechanism by integrating probabilistic metaplasticity into a spiking network trained on an error threshold with low-precision memristor weights. Evaluations of two continual learning benchmarks show that probabilistic metaplasticity consumes  67 parameters and up to two orders of magnitude lower energy during parameter updates compared to an auxiliary memory-based solution while achieving state-of-the-art performance. The proposed model shows potential for energy-efficient continual learning with low-precision emerging devices.

Edge devices operating in dynamic environments critically need the ability to continually learn without catastrophic forgetting. The strict resource constraints in these devices pose a major challenge to achieve this, as continual learning entails memory and computational overhead. Crossbar architectures using memristor devices offer energy efficiency through compute-in-memory and hold promise to address this issue. However, memristors often exhibit low precision and high variability in conductance modulation, rendering them unsuitable for continual learning solutions that require precise modulation of weight magnitude for consolidation. Current approaches fall short to address this challenge directly and rely on auxiliary high-precision memory, leading to frequent memory access, high memory overhead, and energy dissipation. In this research, we propose probabilistic metaplasticity, which consolidates weights by modulating their update probability rather than magnitude. The proposed mechanism eliminates high-precision modification to weight magnitudes and, consequently, the need for auxiliary high-precision memory. We demonstrate the efficacy of the proposed mechanism by integrating probabilistic metaplasticity into a spiking network trained on an error threshold with low-precision memristor weights. Evaluations of continual learning benchmarks show that probabilistic metaplasticity achieves performance equivalent to state-of-the-art continual learning models with high-precision weights while consuming ~ 67% lower memory for additional parameters and up to ~ 60× lower energy during parameter updates compared to an auxiliary memory-based solution. The proposed model shows potential for energy-efficient continual learning with low-precision emerging devices.

Lifelong learning—an agent’s ability to learn throughout its lifetime—is a hallmark of biological learning systems and a central challenge for artificial intelligence (AI). The development of lifelong learning algorithms could lead to a range of novel AI applications, but this will also require the development of appropriate hardware accelerators, particularly if the models are to be deployed on edge platforms, which have strict size, weight and power constraints. Here we explore the design of lifelong learning AI accelerators that are intended for deployment in untethered environments. We identify key desirable capabilities for lifelong learning accelerators and highlight metrics to evaluate such accelerators. We then discuss current edge AI accelerators and explore the future design of lifelong learning accelerators, considering the role that different emerging technologies could play.

The field of neuromorphic computing holds great promise in terms of advancing computing efficiency and capabilities by following brain-inspired principles. However, the rich diversity of techniques employed in neuromorphic research has resulted in a lack of clear standards for benchmarking, hindering effective evaluation of the advantages and strengths of neuromorphic methods compared to traditional deep-learning-based methods. This paper presents a collaborative effort, bringing together members from academia and the industry, to define benchmarks for neuromorphic computing: NeuroBench. The goals of NeuroBench are to be a collaborative, fair, and representative benchmark suite developed by the community, for the community. In this paper, we discuss the challenges associated with benchmarking neuromorphic solutions, and outline the key features of NeuroBench. We believe that NeuroBench will be a significant step towards defining standards that can unify the goals of neuromorphic computing and drive its technological progress. Please visit neurobench.ai for the latest updates on the benchmark tasks and metrics.

Spiking neural network models when deployed in dynamic environments, catastrophically forget previously learned tasks. In this paper, we propose a reconfigurable spiking digital accelerator, which uses activity-dependent metaplasticity to mitigate catastrophic forgetting. The proposed accelerator has a custom low precision dual fixed point representation for network parameters. The custom precision leads to lower quantization error and higher accuracy. We evaluate the proposed accelerator on split-MNIST continual learning benchmark. Analysis shows that representing network parameters with 8-bit dual fixed point numbers reduces the memory footprint compared to 16-bit fixed point numbers, while maintaining comparable continual learning ability.

Metaplasticity, the activity-dependent modification of synaptic plasticity, is an important technique for mitigating catastrophic forgetting in neural networks. Often, continual learning models with metaplasticity require compute-intensive training. In this research, we propose a probabilistic metaplastic synapse with discrete hidden states that alleviates the computational cost. We implement a digital architecture of the network with on-chip training to achieve further power savings. Results show upto ~ 22% and ~ 21% improvement in mean accuracy for Split-MNIST and sequential MNIST-FMNIST benchmarks respectively, compared to previous metaplasticity models. Simulations of the full digital architecture show ~ 53× lower power consumption per weight update with similar accuracy as gradient-based network counterparts.

Recent studies have shown that metaplastic synapses can retain information longer than simple binary synapses and are beneficial for continual learning. In this paper, we explore the multistate metaplastic synapse characteristics in the context of high retention and reception of information. Inherent behavior of a memristor emulating the multistate synapse is employed to capture the metaplastic behavior. An integrated neural network study for learning and memory retention is performed by integrating the synapse in a $5\times3$ crossbar at the circuit level and $128\times128$ network at the architectural level. An on-device training circuitry ensures the dynamic learning in the network. In the $128\times128$ network, it is observed that the number of input patterns the multistate synapse can classify is $\simeq$ 2.1x that of a simple binary synapse model, at a mean accuracy of $\geq$ 75% .

With the onset of on-device learning in neuromorphic systems, there are a requisition for compute-lite learning rules and novel emerging devices that address the memory bottleneck. In this research, we propose a neuromorphic architecture with FeFET synapse arrays and study the efficacy of write schemes for feedback alignment backpropagation algorithm. The proposed architecture is benchmarked for two write programming schemes, sawtooth pulse and incremental pulse. The sawtooth write programming scheme is further simplified for resource efficient training, by sharing the pulse generator with local control circuitry across multiple neurons. When the overall architecture is benchmarked for on-device learning, we observed that both writing schemes result in comparable performance, but the sawtooth is more efficient in terms of power consumption and area.

Unitary operations using linear optics have many applications within the quantum and neuromorphic space. In silicon photonics, using networks of simple beam splitters and phase shifters have proven sufficient to realize large-scale arbitrary unitaries. While this technique has shown success with high fidelity, the grid physically scales with an upper bound of O(n(2)). Consequently, we propose to considerably reduce the footprint by using multimode interference (MMI) devices. In this paper, we investigate the active control of these MMIs and their suitability for approximating traditionally used unitary circuits.

Memristor is a nanoscale device which consumes low power and shows good compatibility with CMOS circuits. It has applications in memory circuits, logic circuits as well as in neuromorphic systems to imitate biological synapses. The use of this device to implement neuron models can improve the scalability of neuromorphic circuits. In this paper a Leaky Integrate and Fire model of neuron is presented by a memristor-CMOS hybrid circuit which requires 16 MOSFETs, 1 memristor and 1 capacitor. The model has been applied in a simple configuration of one neuron driving another. Additionally, it has been used in an associative learning circuit to exhibit functionality. Such successful incorporation of the proposed design in learning networks founds the ground of further expansion and implementation of larger networks using neuron circuits.

A system for monitoring patients for the cold sweat symptom and giving instant signal if they break into sweat has been proposed in this work. A method was devised to create such a system for indicating profuse perspiration by continuous monitoring of Galvanic Skin Resistance. For this purpose a circuit was designed, an algorithm was created and certain steps were followed towards achieving the goal of detecting profuse perspiration. Finally, a system was created that would immediately notify if a person's skin gets moist. Special attention had been given to create the system with easily available components for use in technologically unadvanced nations.