Low Precision RNNs: Quantizing RNNs Without Losing Accuracy
This work addresses efficiency and accuracy trade-offs in RNNs for hardware deployment, but it appears incremental as it builds on existing quantization methods.
The paper tackles the problem of accuracy loss in quantized recurrent neural networks (RNNs) by proposing a quantization approach that increases model size with bit-width reduction, achieving baseline accuracy while maintaining reduced precision and overall model size reduction.
Similar to convolution neural networks, recurrent neural networks (RNNs) typically suffer from over-parameterization. Quantizing bit-widths of weights and activations results in runtime efficiency on hardware, yet it often comes at the cost of reduced accuracy. This paper proposes a quantization approach that increases model size with bit-width reduction. This approach will allow networks to perform at their baseline accuracy while still maintaining the benefits of reduced precision and overall model size reduction.