Toward the long-standing dream of artificial intelligence, two successful solution paths have been paved: 1) neuromorphic computing and 2) deep learning. Recently, they tend to interact for simultaneously achieving biological plausibility and powerful accuracy. However, models from these two domains have to run on distinct substrates, i.e., neuromorphic platforms and deep learning accelerators, respectively. This architectural incompatibility greatly compromises the modeling flexibility and hinders promising interdisciplinary research. To address this issue, we build a unified model description framework and a unified processing architecture (Tianjic), which covers the full stack from software to hardware. By implementing a set of integration and transformation operations, Tianjic is able to support spiking neural networks, biological dynamic neural networks, multilayered perceptron, convolutional neural networks, recurrent neural networks, and so on. A compatible routing infrastructure enables homogeneous and heterogeneous scalability on a decentralized many-core network. Several optimization methods are incorporated, such as resource and data sharing, near-memory processing, compute/access skipping, and intra-/inter-core pipeline, to improve performance and efficiency. We further design streaming mapping schemes for efficient network deployment with a flexible tradeoff between execution throughput and resource overhead. A 28-nm prototype chip is fabricated with >610-GB/s internal memory bandwidth. A variety of benchmarks are evaluated and compared with GPUs and several existing specialized platforms. In summary, the fully unfolded mapping can achieve significantly higher throughput and power efficiency; the semi-folded mapping can save 30x resources while still presenting comparable performance on average. Finally, two hybrid-paradigm examples, a multimodal unmanned bicycle and a hybrid neural network, are demonstrated to show the potential of our unified architecture. This article paves a new way to explore neural computing.

As well known, the huge memory and compute costs of both artificial neural networks (ANNs) and spiking neural networks (SNNs) greatly hinder their deployment on edge devices with high efficiency. Model compression has been proposed as a promising technique to improve the running efficiency via parameter and operation reduction, whereas this technique is mainly practiced in ANNs rather than SNNs. It is interesting to answer how much an SNN model can be compressed without compromising its functionality, where two challenges should be addressed: 1) the accuracy of SNNs is usually sensitive to model compression, which requires an accurate compression methodology and 2) the computation of SNNs is event-driven rather than static, which produces an extra compression dimension on dynamic spikes. To this end, we realize a comprehensive SNN compression through three steps. First, we formulate the connection pruning and weight quantization as a constrained optimization problem. Second, we combine spatiotemporal backpropagation (STBP) and alternating direction method of multipliers (ADMMs) to solve the problem with minimum accuracy loss. Third, we further propose activity regularization to reduce the spike events for fewer active operations. These methods can be applied in either a single way for moderate compression or a joint way for aggressive compression. We define several quantitative metrics to evaluate the compression performance for SNNs. Our methodology is validated in pattern recognition tasks over MNIST, N-MNIST, CIFAR10, and CIFAR100 datasets, where extensive comparisons, analyses, and insights are provided. To the best of our knowledge, this is the first work that studies SNN compression in a comprehensive manner by exploiting all compressible components and achieves better results.

Transformer models achieve excellent results in the fields like natural language processing, computer vision, and bioinformatics. Their large numbers of matrix multiplications (MMs) lead to substantial data movement and computation. Although computing-in-memory (CIM) has proven to be an efficient architecture for MM computation, transformer’s attention mechanism raises new challenges in memory access and computation aspects: the dynamic MM in attention layers causes redundant OFF-chip memory access; Attention layers dominate transformer’s computation and require high precision. Thus, we design a bitline-transpose CIM-based transformer accelerator TranCIM with pipeline/parallel reconfigurable modes. The pipeline mode alleviates off-chip access for attention layers. The parallel mode is used by fully-connected (FC) layers for high parallelism. The full-digital CIM supports INT16 for attention layers and INT8 for FC layers, without analog CIM’s nonideal issues. Moreover, a sparse attention scheduler (SAS) is proposed to reduce attention computation. The fabricated TranCIM chip only consumes 15.59 $\mu \text{J}$ /Token for the bidirectional encoder representations from transformer (BERT)-base model, achieving $12.08\times $ – $36.82\times $ lower energy than prior CIM-based accelerators.

Cloud AI acceleration has drawn great attention in recent years, as big models are becoming a popular trend in deep learning. Cloud AI runs high-efficiency inference, high-accuracy inference and training, in demand of flexible floating-point (FP)/integer (INT) multiply–accumulation (MAC) support Many computing-in-memory (CIM) processors have been proposed for efficient AI acceleration. They usually rely on analog CIM techniques that are only suitable for high-efficiency neural network (NN) inference with low-precision INT MAC support. Since cloud AI demands high efficiency, high accuracy, and high flexibility simultaneously, we propose an innovative architecture reconfigurable digital CIM (ReDCIM) that meets all three requirements. We design the first CIM-based cloud AI processor, ReDCIM, which constructs a unified FP/INT pipeline architecture based on exponent pre-alignment and reconfigurable in-memory accumulation. Bitwise in-memory Booth multiplication is proposed to reduce computation on CIM. The fabricated ReDCIM chip achieves a state-of-the-art energy efficiency of 29.2 TFLOPS/W at BF16 and 36.5 TOPS/W at INT8.

Spiking neural network (SNN) is broadly deployed in neuromorphic devices to emulate brain function. In this context, SNN security becomes important while lacking in-depth investigation. To this end, we target the adversarial attack against SNNs and identify several challenges distinct from the artificial neural network (ANN) attack: 1) current adversarial attack is mainly based on gradient information that presents in a spatiotemporal pattern in SNNs, hard to obtain with conventional backpropagation algorithms; 2) the continuous gradient of the input is incompatible with the binary spiking input during gradient accumulation, hindering the generation of spike-based adversarial examples; and 3) the input gradient can be all-zeros (i.e., vanishing) sometimes due to the zero-dominant derivative of the firing function. Recently, backpropagation through time (BPTT)-inspired learning algorithms are widely introduced into SNNs to improve the performance, which brings the possibility to attack the models accurately given spatiotemporal gradient maps. We propose two approaches to address the above challenges of gradient-input incompatibility and gradient vanishing. Specifically, we design a gradient-to-spike (G2S) converter to convert continuous gradients to ternary ones compatible with spike inputs. Then, we design a restricted spike flipper (RSF) to construct ternary gradients that can randomly flip the spike inputs with a controllable turnover rate, when meeting all-zero gradients. Putting these methods together, we build an adversarial attack methodology for SNNs. Moreover, we analyze the influence of the training loss function and the firing threshold of the penultimate layer on the attack effectiveness. Extensive experiments are conducted to validate our solution. Besides the quantitative analysis of the influence factors, we also compare SNNs and ANNs against adversarial attacks under different attack methods. This work can help reveal what happens in SNN attacks and might stimulate more research on the security of SNN models and neuromorphic devices.