# SONIQ: System-Optimized Noise-Injected Ultra-Low-Precision Quantization with Full-Precision Parity

arXiv: https://arxiv.org/abs/2311.14114 · HTML: https://mlsanity.com/paper/2311.14114/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 42 of its 94 references that are in the map.
- First posted on arXiv 23 Nov 2023.

## Builds on (most cited first, 30 of 42)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration](https://mlsanity.com/paper/2306.00978/) (2023)
- [ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices](https://mlsanity.com/paper/1707.01083/) (2017)
- [Diffusion Models in Vision: A Survey](https://mlsanity.com/paper/2209.04747/) (2022)
- [XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks](https://mlsanity.com/paper/1603.05279/) (2016)
- [Learned Step Size Quantization](https://mlsanity.com/paper/1902.08153/) (2019)
- [In-Datacenter Performance Analysis of a Tensor Processing Unit](https://mlsanity.com/paper/1704.04760/) (2017)
- [BinaryConnect: Training Deep Neural Networks with binary weights during propagations](https://mlsanity.com/paper/1511.00363/) (2015)
- [DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients](https://mlsanity.com/paper/1606.06160/) (2016)
- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [PACT: Parameterized Clipping Activation for Quantized Neural Networks](https://mlsanity.com/paper/1805.06085/) (2018)
- [HAQ: Hardware-Aware Automated Quantization with Mixed Precision](https://mlsanity.com/paper/1811.08886/) (2018)
- [Harnessing the Power of LLMs in Practice: A Survey on ChatGPT and Beyond](https://mlsanity.com/paper/2304.13712/) (2023)
- [Atom: Low-bit Quantization for Efficient and Accurate LLM Serving](https://mlsanity.com/paper/2310.19102/) (2023)
- [Noisy Networks for Exploration](https://mlsanity.com/paper/1706.10295/) (2017)
- [Pruning and Quantization for Deep Neural Network Acceleration: A Survey](https://mlsanity.com/paper/2101.09671/) (2021)
- [Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networks](https://mlsanity.com/paper/1908.05033/) (2019)
- [TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning with Hardware Support for Embeddings](https://mlsanity.com/paper/2304.01433/) (2023)
- [Characterising Bias in Compressed Models](https://mlsanity.com/paper/2010.03058/) (2020)
- [On-Device Training Under 256KB Memory](https://mlsanity.com/paper/2206.15472/) (2022)
- [Training with Quantization Noise for Extreme Model Compression](https://mlsanity.com/paper/2004.07320/) (2020)
- [Bit Fusion: Bit-Level Dynamically Composable Architecture for Accelerating Deep Neural Networks](https://mlsanity.com/paper/1712.01507/) (2017)
- [OliVe: Accelerating Large Language Models via Hardware-friendly Outlier-Victim Pair Quantization](https://mlsanity.com/paper/2304.07493/) (2023)
- [Mixed-Precision Neural Network Quantization via Learned Layer-wise Importance](https://mlsanity.com/paper/2203.08368/) (2022)

## Abstract (arXiv)

Ultra-low-precision inference can sharply reduce memory and latency but often degrades accuracy and relies on specialized hardware. We present SONIQ, a system-optimized, noise-injected quantization framework that learns per-channel mixed precision for both weights and activations while training under the same rules used at inference. By injecting hardware-calibrated quantization noise during training, SONIQ steers models toward the discrete arithmetic used at deployment -- without bespoke runtimes. Across CNNs and Transformers, SONIQ achieves up to 16x and 7x compression, respectively, while matching or exceeding full-precision accuracy. Measured end-to-end, SONIQ delivers up to 7.3x CPU speedup over strong INT8 baselines and up to 6.3x (vector units) / 2.8x (tensor cores) GPU speedup relative to FP16. A practical outcome is that two per-channel precision levels -- one in the 1--4-bit range and one in the 4--8-bit range -- suffice in practice; at inference, each channel selects one of the two, keeping kernels simple and fast. To our knowledge, SONIQ is the first framework to reach or surpass full-precision accuracy under ultra-low (1--4 bits per parameter) regimes while remaining deployable on commodity hardware, narrowing the gap between quantization theory and practical, high-throughput inference.