Skip connections have emerged as a key component of modern convolutional neural networks (CNNs) for computer vision tasks, allowing for the creation of more accurate and deeper models by addressing the vanishing gradient problem. However, the existing implementations of field-programmable gate array (FPGA)-based accelerators for ResNets and MobileNetV2 often experience decreased performance and increased computational latency due to the implementation of skip blocks. This article presents a novel framework for developing deep learning models on FPGAs that focuses on skip connections, with a unique approach to reduce buffering overhead. This results in a more efficient utilization of resources in the implementation of the skip layer. The nn2fpga compiler follows a thorough set of high-level synthesis (HLS) design principles and optimization strategies, exploiting in novel ways standard techniques to effectively map skip connection-based networks into static dataflow accelerators. To maximize throughput and efficiently use the available resources, our compiler employs a fast and effective design space exploration method based on a binary integer programming model which accurately assigns FPGA resources to the network layers, to maximize global throughput under resource constraints and then minimize resources for the achieved maximum throughput. Experimental results on the CIFAR-10 and ImageNet datasets demonstrate substantial gains in throughput ($\mathbf {3\times }$–$\mathbf {7\times }$on the past HLS-based work) for ResNet8, ResNet20, and MobileNetV2 models deployed on various Xilinx FPGA boards. Notably, MobileNetV2 deployed on the ZCU102 achieves a throughput of 2115 frame per second, representing even a 10% speedup over a state-of-the-art highly optimized manual register-transfer level implementation, showing that HLS can actually improve over manual design, thanks to the faster exploration of the design space.

High-level synthesis (HLS) aims at democratizing custom hardware acceleration with highly abstracted software-like descriptions. However, efficient accelerators still require substantial low-level hardware optimizations, defeating the HLS intent. In the context of field-programmable gate arrays, digital signal processors (DSPs) are a crucial resource that typically requires a significant optimization effort for its efficient utilization, especially when used for sub-word vectorization. This work proposes SILVIA, an open-source LLVM transformation pass that automatically identifies superword-level parallelism within an HLS design and exploits it by packing multiple operations, such as additions, multiplications, and multiply-and-adds, into a single DSP. SILVIA is integrated in the flow of the commercial AMD Vitis HLS tool and proves its effectiveness by packing multiple operations on the DSPs without any manual source-code modifications on several diverse state-of-the-art HLS designs such as convolutional neural networks and basic linear algebra subprograms accelerators, reducing the DSP utilization for additions by 70% and for multiplications and multiply-and-adds by 50% on average.

Low-power energy-efflcient subgraph isomorphism (LESS) is an open-source field-programmable gate array-only low-memory sub graph matching solver designed for energy efficiency. Depending on the input datagraph, the energy consumption of LESS, averaged on different diverse queries, is up to 38x and 93x lower than CPU and GPU solvers respectively.