We introduce Bayesian Bits, a practical method for joint mixed precision quantization and pruning through gradient based optimization. Bayesian Bits employs a novel decomposition of the quantization operation, which sequentially considers doubling the bit width. At each new bit width, the residual error between the full precision value and the previously rounded value is quantized. We then decide whether or not to add this quantized residual error for a higher effective bit width and lower quantization noise. By starting with a power-of-two bit width, this decomposition will always produce hardware-friendly configurations, and through an additional 0-bit option, serves as a unified view of pruning and quantization. Bayesian Bits then introduces learnable stochastic gates, which collectively control the bit width of the given tensor. As a result, we can obtain low bit solutions by performing approximate inference over the gates, with prior distributions that encourage most of them to be switched off. We experimentally validate our proposed method on several benchmark datasets and show that we can learn pruned, mixed precision networks that provide a better trade-off between accuracy and efficiency than their static bit width equivalents.
Convolutional Neural Networks (CNNs) suffer from different issues, such as computational complexity and the number of parameters. In recent years pruning techniques are employed to reduce the number of operations and model size in CNNs. Different pruning methods are proposed, which are based on pruning the connections, channels, and filters. Various techniques and tricks accompany pruning methods, and there is not a unifying framework to model all the pruning methods. In this paper pruning methods are investigated, and a general model which is contained the majority of pruning techniques is pr…
We propose the position-based scaled gradient (PSG) that scales the gradient depending on the position of a weight vector to make it more compression-friendly. First, we theoretically show that applying PSG to the standard gradient descent (GD), which is called PSGD, is equivalent to the GD in the warped weight space, a space made by warping the original weight space via an appropriately designed invertible function. Second, we empirically show that PSG acting as a regularizer to a weight vector is favorable for model compression domains such as quantization and pruning. PSG reduces the gap be…
Unifying Synergies between Self-supervised Learning and Dynamic Computation
Insight Centre for Data Analytics（洞察数据分析中心）；Dublin City University（都柏林城市大学）；School of Computing（计算机学院）；School of Electronic Engineering（电子工程学院）；Xperi Corporation（克佩里公司）
Friedrich-Alexander University Erlangen-Nürnberg (FAU)（埃尔朗根-纽伦堡大学）
For many applications, utilizing DNNs (Deep Neural Networks) requires their implementation on a target architecture in an optimized manner concerning energy consumption, memory requirement, throughput, etc. DNN compression is used to reduce the memory footprint and complexity of a DNN before its deployment on hardware. Recent efforts to understand and explain AI (Artificial Intelligence) methods have led to a new research area, termed as explainable AI. Explainable AI methods allow us to understand better the inner working of DNNs, such as the importance of different neurons and features. The …
Large deep neural network (DNN) models pose the key challenge to energy efficiency due to the significantly higher energy consumption of off-chip DRAM accesses than arithmetic or SRAM operations. It motivates the intensive research on model compression with two main approaches. Weight pruning leverages the redundancy in the number of weights and can be performed in a non-structured, which has higher flexibility and pruning rate but incurs index accesses due to irregular weights, or structured manner, which preserves the full matrix structure with lower pruning rate. Weight quantization leverag…
Cranberry-Lemon University（蔓越莓柠檬大学）；University of the Witwatersrand（金山大学）
Deep Neural Networks(DNNs) have many parameters and activation data, and these both are expensive to implement. One method to reduce the size of the DNN is to quantize the pre-trained model by using a low-bit expression for weights and activations, using fine-tuning to recover the drop in accuracy. However, it is generally difficult to train neural networks which use low-bit expressions. One reason is that the weights in the middle layer of the DNN have a wide dynamic range and so when quantizing the wide dynamic range into a few bits, the step size becomes large, which leads to a large quanti…
Neural networks are both computationally intensive and memory intensive, making them difficult to deploy on embedded systems with limited hardware resources. To address this limitation, we introduce "deep compression", a three stage pipeline: pruning, trained quantization and Huffman coding, that work together to reduce the storage requirement of neural networks by 35x to 49x without affecting their accuracy. Our method first prunes the network by learning only the important connections. Next, we quantize the weights to enforce weight sharing, finally, we apply Huffman coding. After the first …
Unifying Depth and Width Pruning for LLMs via Binary Knapsack Optimization
Yardi School of Artificial Intelligence, Indian Institute of Technology Delhi（德里印度理工学院亚迪人工智能学院）；Indian Institute of Technology Delhi（德里印度理工学院）；Microsoft Research (MSR) India（微软研究院印度分部）
BTC-LLM: Efficient Sub-1-Bit LLM Quantization via Learnable Transformation and Binary Codebook
The Hong Kong University of Science and Technology（香港科技大学）；City University of Hong Kong（香港城市大学）；University of Electronic Science and Technology of China（电子科技大学）
Model compression techniques, such as pruning and quantization, are becoming increasingly important to reduce the memory footprints and the amount of computations. Despite model size reduction, achieving performance enhancement on devices is, however, still challenging mainly due to the irregular representations of sparse matrix formats. This paper proposes a new weight representation scheme for Sparse Quantized Neural Networks, specifically achieved by fine-grained and unstructured pruning method. The representation is encrypted in a structured regular format, which can be efficiently decoded…
Binary networks are extremely efficient as they use only two symbols to define the network: $\{+1,-1\}$. One can make the prior distribution of these symbols a design choice. The recent IR-Net of Qin et al. argues that imposing a Bernoulli distribution with equal priors (equal bit ratios) over the binary weights leads to maximum entropy and thus minimizes information loss. However, prior work cannot precisely control the binary weight distribution during training, and therefore cannot guarantee maximum entropy. Here, we show that quantizing using optimal transport can guarantee any bit ratio, …
Model compression has gained a lot of attention due to its ability to reduce hardware resource requirements significantly while maintaining accuracy of DNNs. Model compression is especially useful for memory-intensive recurrent neural networks because smaller memory footprint is crucial not only for reducing storage requirement but also for fast inference operations. Quantization is known to be an effective model compression method and researchers are interested in minimizing the number of bits to represent parameters. In this work, we introduce an iterative technique to apply quantization, pr…
Columbia University（哥伦比亚大学）；Peking University（北京大学）；Deepwise AI Lab（深度智耀人工智能实验室）；The University of Hong Kong（香港大学）
This paper presents a novel network compression framework Kernel Quantization (KQ), targeting to efficiently convert any pre-trained full-precision convolutional neural network (CNN) model into a low-precision version without significant performance loss. Unlike existing methods struggling with weight bit-length, KQ has the potential in improving the compression ratio by considering the convolution kernel as the quantization unit. Inspired by the evolution from weight pruning to filter pruning, we propose to quantize in both kernel and weight level. Instead of representing each weight paramete…
MSQ: Memory-Efficient Bit Sparsification Quantization
Department of Electrical and Computer Engineering, Sungkyunkwan University（全州大学电子与计算机工程系）；Department of Electrical and Computer Engineering, University of Arizona（亚利桑那大学电子与计算机工程系）
