The growing popularity of edgeAI requires novel solutions to support the deployment of compute-intense algorithms in embedded devices. In this article, we advocate for a holistic approach, where application-level transformations are jointly conceived with dedicated hardware platforms. We embody such a stance in a strategy that employs ensemble-based algorithmic transformations to increase robustness and accuracy in Convolutional Neural Networks (CNNs), enabling the aggressive quantization of weights and activations. Opportunities offered by algorithmic optimizations are then harnessed in domain-specific hardware solutions, such as the use of multiple ultra-low-power processing cores, the provision of shared acceleration resources, the presence of independently power-managed memory banks, and voltage scaling to ultra-low levels, greatly reducing (up to 60% in our experiments) energy requirements. Furthermore, we show that aggressive quantization schemes can be leveraged to perform efficient computations directly in memory banks, adopting in-memory computing solutions. We showcase that the combination of parallel in-memory execution and aggressive quantization leads to more than 70% energy and latency gains compared to baseline implementations.

To reduce energy consumption, it is possible to operate embedded systems at sub-nominal conditions (e.g., reduced voltage, limited eDRAM refresh rate) that can introduce bit errors in their memories. These errors can affect the stored values of convolutional neural network (CNN) weights and activations, compromising their accuracy. In this article, we introduce Embedded Ensemble CNNs (E ² CNNs), our architectural design methodology to conceive ensembles of convolutional neural networks to improve robustness against memory errors compared to a single-instance network. Ensembles of CNNs have been previously proposed to increase accuracy at the cost of replicating similar or different architectures. Unfortunately, state-of-the-art (SoA) ensembles do not suit well embedded systems, in which memory and processing constraints limit the number of deployable models. Our proposed architecture solves that limitation applying SoA compression methods to produce an ensemble with the same memory requirements of the original architecture, but with improved error robustness. Then, as part of our new E ² CNNs design methodology, we propose a heuristic method to automate the design of the voter-based ensemble architecture that maximizes accuracy for the expected memory error rate while bounding the design effort. To evaluate the robustness of E ² CNNs for different error types and densities, and their ability to achieve energy savings, we propose three error models that simulate the behavior of SRAM and eDRAM operating at sub-nominal conditions. Our results show that E ² CNNs achieves energy savings of up to 80 percent for LeNet-5, 90 percent for AlexNet, 60 percent for GoogLeNet, 60 percent for MobileNet and 60 percent for an optimized industrial CNN, while minimizing the impact on accuracy. Furthermore, the memory size can be decreased up to 54 percent by reducing the number of members in the ensemble, with a more limited impact on the original accuracy than obtained through pruning alone.

By supporting the access of multiple memory words at the same time, Bit-line Computing (BC) architectures allow the parallel execution of bit-wise operations in-memory. At the array periphery, arithmetic operations are then derived with little additional overhead. Such a paradigm opens novel opportunities for Artificial Intelligence (AI) at the edge, thanks to the massive parallelism inherent in memory arrays and the extreme energy efficiency of computing in-situ, hence avoiding data transfers. Previous works have shown that BC brings disruptive efficiency gains when targeting AI workloads, a key metric in the context of emerging edge AI scenarios. This manuscript builds on these findings by proposing an end-to-end framework that leverages BC-specific optimizations to enable high parallelism and aggressive compression of AI models. Our approach is supported by a novel hardware module performing real-time decoding, as well as new algorithms to enable BC-friendly model compression. Our hardware/software approach results in a 91% energy savings (for a 1% accuracy degradation constraint) regarding state-of-the-art BC computing approaches.

When arranged in a crossbar configuration, resistive memory devices can be used to execute Matrix-Vector Multiplications (MVMs), the most dominant operation of many Machine Learning (ML) algorithms, in constant time complexity. Nonetheless, when performing computations in the analog domain, novel challenges are introduced in terms of arithmetic precision and stochasticity, due to non-ideal circuit and device behaviour. Moreover, these non-idealities have a temporal dimension, resulting in a degrading application accuracy over time. Facing these challenges, we propose a novel framework, named LionHeart, to obtain hybrid analog-digital mappings to execute Deep Learning (DL) inference workloads using heterogeneous accelerators. The accuracy-constrained mappings derived by LionHeart showcase, across different Convolutional Neural Networks (CNNs) and one transformer-based network, high accuracy and potential for speedup. The results of the full system simulations highlight runtime reductions and energy efficiency gains that exceed 6×, with a user-defined accuracy threshold for a fully digital floating point implementation.

Mass spectrometry (MS) is essential for proteomics and metabolomics but faces impending challenges in efficiently processing the vast volumes of data. This paper introduces SpecPCM, an in-memory computing (IMC) accelerator designed to achieve substantial improvements in energy and delay efficiency for both MS spectral clustering and database (DB) search. SpecPCM employs analog processing with low-voltage swing and utilizes recently introduced phase change memory (PCM) devices based on superlattice materials, optimized for low-voltage and low-power programming. Our approach integrates contributions across multiple levels: application, algorithm, circuit, device, and instruction sets. We leverage a robust hyperdimensional computing (HD) algorithm with a novel dimension-packing method and develop specialized hardware for the end-to-end MS pipeline to overcome the non-ideal behavior of PCM devices. We further optimize multi-level PCM devices for different tasks by using different materials. We also perform a comprehensive design exploration to improve energy and delay efficiency while maintaining accuracy, exploring various combinations of hardware and software parameters controlled by the instruction set architecture (ISA). SpecPCM, with up to three bits per cell, achieves speedups of up to 82× and 143× for MS clustering and DB search tasks, respectively, along with a four-orders-of-magnitude improvement in energy efficiency compared with state-of-the-art CPU/GPU tools.