SparseLUT: Sparse Connectivity Optimization for Lookup Table-based Deep Neural Networks
This work addresses the challenge of balancing latency, power, and resource usage for DNNs on edge devices, representing an incremental improvement over existing LUT-based methods.
The paper tackled the problem of optimizing sparse connectivity in Lookup Table-based deep neural networks for deployment on resource-constrained edge devices like FPGAs, resulting in consistent accuracy improvements such as up to 2.13% on MNIST and 0.94% on Jet Substructure Classification compared to random sparsity without hardware overhead.
The deployment of deep neural networks (DNNs) on resource-constrained edge devices such as field-programmable gate arrays (FPGAs) requires a careful balance of latency, power, and resource usage while maintaining high accuracy. Existing Lookup Table (LUT)-based DNNs, including LogicNets, PolyLUT, PolyLUT-Add, and NeuraLUT, exploit native FPGA resources with random sparse connectivity. This paper introduces SparseLUT, a connectivity-centric training technique tailored for LUT-based DNNs. SparseLUT leverages a non-greedy training strategy that prioritizes the pruning of less significant connections and strategically regrows alternative ones, resulting in efficient convergence to the target sparsity. Experimental results show consistent accuracy improvements across benchmarks, including up to a 2.13\% increase on MNIST and a 0.94\% improvement for Jet Substructure Classification compared to random sparsity. This is done without any hardware overhead and achieves state-of-the-art results for LUT-based DNNs.