UNIQ: Uniform Noise Injection for Non-Uniform Quantization of Neural Networks
This work addresses the problem of efficient neural network deployment for resource-constrained applications, though it is incremental as it builds on existing quantization methods.
The paper tackles neural network quantization by introducing a non-uniform quantizer that adapts to parameter distributions, showing advantages in low computational budgets when measured by bit-operations.
We present a novel method for neural network quantization that emulates a non-uniform $k$-quantile quantizer, which adapts to the distribution of the quantized parameters. Our approach provides a novel alternative to the existing uniform quantization techniques for neural networks. We suggest to compare the results as a function of the bit-operations (BOPS) performed, assuming a look-up table availability for the non-uniform case. In this setup, we show the advantages of our strategy in the low computational budget regime. While the proposed solution is harder to implement in hardware, we believe it sets a basis for new alternatives to neural networks quantization.