Differentiable Weightless Neural Networks
This work provides a pioneering solution for edge-compatible high-throughput neural networks, addressing efficiency and deployment challenges in resource-constrained environments.
The paper tackles the problem of enabling neural networks for edge computing by introducing Differentiable Weightless Neural Networks (DWNs), which achieve superior latency, throughput, energy efficiency, and model area compared to state-of-the-art solutions in FPGA-based accelerators, outperform XGBoost in accuracy under memory constraints on microcontrollers, and consistently beat small models in accuracy and hardware area on ultra-low-cost chips.
We introduce the Differentiable Weightless Neural Network (DWN), a model based on interconnected lookup tables. Training of DWNs is enabled by a novel Extended Finite Difference technique for approximate differentiation of binary values. We propose Learnable Mapping, Learnable Reduction, and Spectral Regularization to further improve the accuracy and efficiency of these models. We evaluate DWNs in three edge computing contexts: (1) an FPGA-based hardware accelerator, where they demonstrate superior latency, throughput, energy efficiency, and model area compared to state-of-the-art solutions, (2) a low-power microcontroller, where they achieve preferable accuracy to XGBoost while subject to stringent memory constraints, and (3) ultra-low-cost chips, where they consistently outperform small models in both accuracy and projected hardware area. DWNs also compare favorably against leading approaches for tabular datasets, with higher average rank. Overall, our work positions DWNs as a pioneering solution for edge-compatible high-throughput neural networks.