arXiv subjects

Anbang Yao

Publications and source records attributed to Anbang Yao.

At least 19 recordsLinked to original sources

arXiv2026

Attend to Your Own Thoughts: Breaking the Barrier for Post-Training Quantization of Reasoning LLMs through the Lens of 1.58-Bit Quantization

We propose ScaleQ-1.58, a scalable ternary post-training quantization (PTQ) framework for reasoning LLMs. Its core insight stems from an empirical finding: although modern LLMs are typically trained to exhibit chain-of-thought reasoning capabilities, in the PTQ regime, even the latest CAT-Q method based on learning-based differentiable ternarization still leads to performance collapse on challenging mathematics and coding tasks when using conventional calibration schemes that ignore the model's reasoning process. Driven by this finding, we introduce a simple calibration approach, Attend to Your Own Thoughts (AYOT), where reasoning traces and final answers generated by the pre-trained high-precision target LLM on a proper set of calibration samples are used as the context input during the ternarization process, along with the corresponding questions. ScaleQ-1.58 is formed by simply integrating AYOT with CAT-Q, which demonstrates several scaling properties: (1) with only 4M calibration tokens, Qwen3-1.7B ternarized by ScaleQ-1.58 reaches over 90.52% of the performance of the prior best BitNet b1.58 2B4T averaged over 4 mathematics and coding tasks, and our ternary Qwen3-4B shows an absolute gain of 8.97%, while requiring 1,000,000x fewer calibration tokens for quantization; (2) ScaleQ-1.58 generalizes well to both dense and MoE architectures, with performance improving as model scale increases (up to 235B parameters); (3) ScaleQ-1.58 demonstrates strong generalization across tasks of varying difficulty levels, including mathematics, coding and scientific logic reasoning, as well as commonsense reasoning and basic language generation; (4) its performance continues to improve as the number of calibration tokens increases. Notably, AYOT also exhibits strong generalization ability across other quantization bit-widths. Code will be available at https://github.com/IntelChina-AI/BitTern.

cs.CL↗

arXiv2026

CAT-Q: Cost-efficient and Accurate Ternary Quantization for LLMs

In this paper, we present CAT-Q, Cost-efficient and Accurate Ternary Quantization, for compressing and accelerating LLMs. Unlike existing state-of-the-art ternary quantization methods that rely on data-intensive and costly quantization-aware training to mitigate severe performance degradation, CAT-Q is a simple yet effective post-training quantization scheme that is readily applicable to LLMs with diverse architectures and model sizes. It has two key components, learnable modulation (LM) and softened ternarization (ST), which are coupled from an optimization perspective. LM leverages a composition of learnable factors to modulate the distribution of pre-trained high-precision weights and the ternary threshold, making them less sensitive to ternarization. ST further introduces a differentiable transition function to guide the ternarization process toward stable convergence. We show that, for pre-trained LLMs with 1.7B to 8B parameters, CAT-Q can efficiently quantize them into ternary models using only 512 calibration samples, while achieving superior performance than the seminal BitNet 1.58-bit v1 and v2 families (with 1.3B to 7B parameters) trained with 100B tokens, yielding about a 100,000X reduction in training tokens. Moreover, we show for the first time that CAT-Q can quantize much larger pre-trained LLMs having 14B to 235B parameters into leading ternary models within just 8 to 60 hours on 8 A100-80GB GPUs. Code is available at https://github.com/IntelChina-AI/BitTern.

cs.CL↗

arXiv2026

Chain-of-Models Pre-Training: Rethinking Training Acceleration of Vision Foundation Models

In this paper, we present Chain-of-Models Pre-Training (CoM-PT), a novel performance-lossless training acceleration method for vision foundation models (VFMs). This approach fundamentally differs from existing acceleration methods in its core motivation: rather than optimizing each model individually, CoM-PT is designed to accelerate the training pipeline at the model family level, scaling efficiently as the model family expands. Specifically, CoM-PT establishes a pre-training sequence for the model family, arranged in ascending order of model size, called model chain. In this chain, only the smallest model undergoes standard individual pre-training, while the other models are efficiently trained through sequential inverse knowledge transfer from their smaller predecessors by jointly reusing the knowledge in the parameter space and the feature space. As a result, CoM-PT enables all models to achieve performance that is mostly superior to standard individual training while significantly reducing training cost, and this is extensively validated across 45 datasets spanning zero-shot and fine-tuning tasks. Notably, its efficient scaling property yields a remarkable phenomenon: training more models even results in higher efficiency. For instance, when pre-training on CC3M: i) given ViT-L as the largest model, progressively prepending smaller models to the model chain reduces computational complexity by up to 72%; ii) within a fixed model size range, as the VFM family scales across 3, 4, and 7 models, the acceleration ratio of CoM-PT exhibits a striking leap: from 4.13X to 5.68X and 7.09X. Since CoM-PT is naturally agnostic to specific pre-training paradigms, we open-source the code to spur further extensions in more computationally intensive scenarios, such as large language model pre-training.

cs.CV↗

arXiv2026

SliderQuant: Accurate Post-Training Quantization for LLMs

In this paper, we address post-training quantization (PTQ) for large language models (LLMs) from an overlooked perspective: given a pre-trained high-precision LLM, the predominant sequential quantization framework treats different layers equally, but this may be not optimal in challenging bit-width settings. We empirically study the quantization impact of different layers on model accuracy, and observe that: (1) shallow/deep layers are usually more sensitive to quantization than intermediate layers; (2) among shallow/deep layers, the most sensitive one is the first/last layer, which exhibits significantly larger quantization error than others. These empirical observations imply that the quantization design for different layers of LLMs is required on multiple levels instead of a single level shared to all layers. Motivated by this, we propose a new PTQ framework termed Sliding-layer Quantization (SliderQuant) that relies on a simple adaptive sliding quantization concept facilitated by few learnable parameters. The base component of SliderQuant is called inter-layer sliding quantization, which incorporates three types of novel sliding window designs tailored for addressing the varying quantization sensitivity of shallow, intermediate and deep layers. The other component is called intra-layer sliding quantization that leverages an incremental strategy to quantize each window. As a result, SliderQuant has a strong ability to reduce quantization errors across layers. Extensive experiments on basic language generation, zero-shot commonsense reasoning and challenging math and code tasks with various LLMs, including Llama/Llama2/Llama3/Qwen2.5 model families, DeepSeek-R1 distilled models and large MoE models, show that our method outperforms existing PTQ methods (including the latest PTQ methods using rotation transformations) for both weight-only quantization and weight-activation quantization.

cs.AI↗

arXiv2025

Morse: Dual-Sampling for Lossless Acceleration of Diffusion Models

In this paper, we present Morse, a simple dual-sampling framework for accelerating diffusion models losslessly. The key insight of Morse is to reformulate the iterative generation (from noise to data) process via taking advantage of fast jump sampling and adaptive residual feedback strategies. Specifically, Morse involves two models called Dash and Dot that interact with each other. The Dash model is just the pre-trained diffusion model of any type, but operates in a jump sampling regime, creating sufficient space for sampling efficiency improvement. The Dot model is significantly faster than the Dash model, which is learnt to generate residual feedback conditioned on the observations at the current jump sampling point on the trajectory of the Dash model, lifting the noise estimate to easily match the next-step estimate of the Dash model without jump sampling. By chaining the outputs of the Dash and Dot models run in a time-interleaved fashion, Morse exhibits the merit of flexibly attaining desired image generation performance while improving overall runtime efficiency. With our proposed weight sharing strategy between the Dash and Dot models, Morse is efficient for training and inference. Our method shows a lossless speedup of 1.78X to 3.31X on average over a wide range of sampling step budgets relative to 9 baseline diffusion models on 6 image generation tasks. Furthermore, we show that our method can be also generalized to improve the Latent Consistency Model (LCM-SDXL, which is already accelerated with consistency distillation technique) tailored for few-step text-to-image synthesis. The code and models are available at https://github.com/deep-optimization/Morse.

cs.GR↗

arXiv2024

ScaleKD: Strong Vision Transformers Could Be Excellent Teachers

In this paper, we question if well pre-trained vision transformer (ViT) models could be used as teachers that exhibit scalable properties to advance cross architecture knowledge distillation (KD) research, in the context of using large-scale datasets for evaluation. To make this possible, our analysis underlines the importance of seeking effective strategies to align (1) feature computing paradigm differences, (2) model scale differences, and (3) knowledge density differences. By combining three coupled components namely cross attention projector, dual-view feature mimicking and teacher parameter perception tailored to address the above problems, we present a simple and effective KD method, called ScaleKD. Our method can train student backbones that span across a variety of convolutional neural network (CNN), multi-layer perceptron (MLP), and ViT architectures on image classification datasets, achieving state-of-the-art distillation performance. For instance, taking a well pre-trained Swin-L as the teacher model, our method gets 75.15%|82.03%|84.16%|78.63%|81.96%|83.93%|83.80%|85.53% top-1 accuracies for MobileNet-V1|ResNet-50|ConvNeXt-T|Mixer-S/16|Mixer-B/16|ViT-S/16|Swin-T|ViT-B/16 models trained on ImageNet-1K dataset from scratch, showing 3.05%|3.39%|2.02%|4.61%|5.52%|4.03%|2.62%|3.73% absolute gains to the individually trained counterparts. Intriguingly, when scaling up the size of teacher models or their pre-training datasets, our method showcases the desired scalable properties, bringing increasingly larger gains to student models. The student backbones trained by our method transfer well on downstream MS-COCO and ADE20K datasets. More importantly, our method could be used as a more efficient alternative to the time-intensive pre-training paradigm for any target student model if a strong pre-trained ViT is available, reducing the amount of viewed training samples up to 195x.

cs.CV↗

arXiv2024

Small Scale Data-Free Knowledge Distillation

Data-free knowledge distillation is able to utilize the knowledge learned by a large teacher network to augment the training of a smaller student network without accessing the original training data, avoiding privacy, security, and proprietary risks in real applications. In this line of research, existing methods typically follow an inversion-and-distillation paradigm in which a generative adversarial network on-the-fly trained with the guidance of the pre-trained teacher network is used to synthesize a large-scale sample set for knowledge distillation. In this paper, we reexamine this common data-free knowledge distillation paradigm, showing that there is considerable room to improve the overall training efficiency through a lens of ``small-scale inverted data for knowledge distillation". In light of three empirical observations indicating the importance of how to balance class distributions in terms of synthetic sample diversity and difficulty during both data inversion and distillation processes, we propose Small Scale Data-free Knowledge Distillation SSD-KD. In formulation, SSD-KD introduces a modulating function to balance synthetic samples and a priority sampling function to select proper samples, facilitated by a dynamic replay buffer and a reinforcement learning strategy. As a result, SSD-KD can perform distillation training conditioned on an extremely small scale of synthetic samples (e.g., 10X less than the original training data scale), making the overall training efficiency one or two orders of magnitude faster than many mainstream methods while retaining superior or competitive model performance, as demonstrated on popular image classification and semantic segmentation benchmarks. The code is available at https://github.com/OSVAI/SSD-KD.

cs.CV↗

arXiv2024

KernelWarehouse: Rethinking the Design of Dynamic Convolution

Dynamic convolution learns a linear mixture of n static kernels weighted with their input-dependent attentions, demonstrating superior performance than normal convolution. However, it increases the number of convolutional parameters by n times, and thus is not parameter efficient. This leads to no research progress that can allow researchers to explore the setting n>100 (an order of magnitude larger than the typical setting n<10) for pushing forward the performance boundary of dynamic convolution while enjoying parameter efficiency. To fill this gap, in this paper, we propose KernelWarehouse, a more general form of dynamic convolution, which redefines the basic concepts of ``kernels", ``assembling kernels" and ``attention function" through the lens of exploiting convolutional parameter dependencies within the same layer and across neighboring layers of a ConvNet. We testify the effectiveness of KernelWarehouse on ImageNet and MS-COCO datasets using various ConvNet architectures. Intriguingly, KernelWarehouse is also applicable to Vision Transformers, and it can even reduce the model size of a backbone while improving the model accuracy. For instance, KernelWarehouse (n=4) achieves 5.61%|3.90%|4.38% absolute top-1 accuracy gain on the ResNet18|MobileNetV2|DeiT-Tiny backbone, and KernelWarehouse (n=1/4) with 65.10% model size reduction still achieves 2.29% gain on the ResNet18 backbone. The code and models are available at https://github.com/OSVAI/KernelWarehouse.

cs.CV↗

arXiv2024

NOAH: Learning Pairwise Object Category Attentions for Image Classification

A modern deep neural network (DNN) for image classification tasks typically consists of two parts: a backbone for feature extraction, and a head for feature encoding and class predication. We observe that the head structures of mainstream DNNs adopt a similar feature encoding pipeline, exploiting global feature dependencies while disregarding local ones. In this paper, we revisit the feature encoding problem, and propose Non-glObal Attentive Head (NOAH) that relies on a new form of dot-product attention called pairwise object category attention (POCA), efficiently exploiting spatially dense category-specific attentions to augment classification performance. NOAH introduces a neat combination of feature split, transform and merge operations to learn POCAs at local to global scales. As a drop-in design, NOAH can be easily used to replace existing heads of various types of DNNs, improving classification performance while maintaining similar model efficiency. We validate the effectiveness of NOAH on ImageNet classification benchmark with 25 DNN architectures spanning convolutional neural networks, vision transformers and multi-layer perceptrons. In general, NOAH is able to significantly improve the performance of lightweight DNNs, e.g., showing 3.14\%|5.3\%|1.9\% top-1 accuracy improvement to MobileNetV2 (0.5x)|Deit-Tiny (0.5x)|gMLP-Tiny (0.5x). NOAH also generalizes well when applied to medium-size and large-size DNNs. We further show that NOAH retains its efficacy on other popular multi-class and multi-label image classification benchmarks as well as in different training regimes, e.g., showing 3.6\%|1.1\% mAP improvement to large ResNet101|ViT-Large on MS-COCO dataset. Project page: https://github.com/OSVAI/NOAH.

cs.CV↗

arXiv2023

Augmentation-Free Dense Contrastive Knowledge Distillation for Efficient Semantic Segmentation

In recent years, knowledge distillation methods based on contrastive learning have achieved promising results on image classification and object detection tasks. However, in this line of research, we note that less attention is paid to semantic segmentation. Existing methods heavily rely on data augmentation and memory buffer, which entail high computational resource demands when applying them to handle semantic segmentation that requires to preserve high-resolution feature maps for making dense pixel-wise predictions. In order to address this problem, we present Augmentation-free Dense Contrastive Knowledge Distillation (Af-DCD), a new contrastive distillation learning paradigm to train compact and accurate deep neural networks for semantic segmentation applications. Af-DCD leverages a masked feature mimicking strategy, and formulates a novel contrastive learning loss via taking advantage of tactful feature partitions across both channel and spatial dimensions, allowing to effectively transfer dense and structured local knowledge learnt by the teacher model to a target student model while maintaining training efficiency. Extensive experiments on five mainstream benchmarks with various teacher-student network pairs demonstrate the effectiveness of our approach. For instance, the DeepLabV3-Res18|DeepLabV3-MBV2 model trained by Af-DCD reaches 77.03%|76.38% mIOU on Cityscapes dataset when choosing DeepLabV3-Res101 as the teacher, setting new performance records. Besides that, Af-DCD achieves an absolute mIOU improvement of 3.26%|3.04%|2.75%|2.30%|1.42% compared with individually trained counterpart on Cityscapes|Pascal VOC|Camvid|ADE20K|COCO-Stuff-164K. Code is available at https://github.com/OSVAI/Af-DCD

cs.CV↗

arXiv2023

KernelWarehouse: Towards Parameter-Efficient Dynamic Convolution

Dynamic convolution learns a linear mixture of $n$ static kernels weighted with their sample-dependent attentions, demonstrating superior performance compared to normal convolution. However, existing designs are parameter-inefficient: they increase the number of convolutional parameters by $n$ times. This and the optimization difficulty lead to no research progress in dynamic convolution that can allow us to use a significant large value of $n$ (e.g., $n>100$ instead of typical setting $n<10$) to push forward the performance boundary. In this paper, we propose $KernelWarehouse$, a more general form of dynamic convolution, which can strike a favorable trade-off between parameter efficiency and representation power. Its key idea is to redefine the basic concepts of "$kernels$" and "$assembling$ $kernels$" in dynamic convolution from the perspective of reducing kernel dimension and increasing kernel number significantly. In principle, KernelWarehouse enhances convolutional parameter dependencies within the same layer and across successive layers via tactful kernel partition and warehouse sharing, yielding a high degree of freedom to fit a desired parameter budget. We validate our method on ImageNet and MS-COCO datasets with different ConvNet architectures, and show that it attains state-of-the-art results. For instance, the ResNet18|ResNet50|MobileNetV2|ConvNeXt-Tiny model trained with KernelWarehouse on ImageNet reaches 76.05%|81.05%|75.52%|82.51% top-1 accuracy. Thanks to its flexible design, KernelWarehouse can even reduce the model size of a ConvNet while improving the accuracy, e.g., our ResNet18 model with 36.45%|65.10% parameter reduction to the baseline shows 2.89%|2.29% absolute improvement to top-1 accuracy.

cs.CV↗

arXiv2023

Ske2Grid: Skeleton-to-Grid Representation Learning for Action Recognition

This paper presents Ske2Grid, a new representation learning framework for improved skeleton-based action recognition. In Ske2Grid, we define a regular convolution operation upon a novel grid representation of human skeleton, which is a compact image-like grid patch constructed and learned through three novel designs. Specifically, we propose a graph-node index transform (GIT) to construct a regular grid patch through assigning the nodes in the skeleton graph one by one to the desired grid cells. To ensure that GIT is a bijection and enrich the expressiveness of the grid representation, an up-sampling transform (UPT) is learned to interpolate the skeleton graph nodes for filling the grid patch to the full. To resolve the problem when the one-step UPT is aggressive and further exploit the representation capability of the grid patch with increasing spatial size, a progressive learning strategy (PLS) is proposed which decouples the UPT into multiple steps and aligns them to multiple paired GITs through a compact cascaded design learned progressively. We construct networks upon prevailing graph convolution networks and conduct experiments on six mainstream skeleton-based action recognition datasets. Experiments show that our Ske2Grid significantly outperforms existing GCN-based solutions under different benchmark settings, without bells and whistles. Code and models are available at https://github.com/OSVAI/Ske2Grid

cs.CV↗

arXiv2023

NORM: Knowledge Distillation via N-to-One Representation Matching

Existing feature distillation methods commonly adopt the One-to-one Representation Matching between any pre-selected teacher-student layer pair. In this paper, we present N-to-One Representation (NORM), a new two-stage knowledge distillation method, which relies on a simple Feature Transform (FT) module consisting of two linear layers. In view of preserving the intact information learnt by the teacher network, during training, our FT module is merely inserted after the last convolutional layer of the student network. The first linear layer projects the student representation to a feature space having N times feature channels than the teacher representation from the last convolutional layer, and the second linear layer contracts the expanded output back to the original feature space. By sequentially splitting the expanded student representation into N non-overlapping feature segments having the same number of feature channels as the teacher's, they can be readily forced to approximate the intact teacher representation simultaneously, formulating a novel many-to-one representation matching mechanism conditioned on a single teacher-student layer pair. After training, such an FT module will be naturally merged into the subsequent fully connected layer thanks to its linear property, introducing no extra parameters or architectural modifications to the student network at inference. Extensive experiments on different visual recognition benchmarks demonstrate the leading performance of our method. For instance, the ResNet18|MobileNet|ResNet50-1/4 model trained by NORM reaches 72.14%|74.26%|68.03% top-1 accuracy on the ImageNet dataset when using a pre-trained ResNet34|ResNet50|ResNet50 model as the teacher, achieving an absolute improvement of 2.01%|4.63%|3.03% against the individually trained counterpart. Code is available at https://github.com/OSVAI/NORM

cs.CV↗

arXiv2023

Compacting Binary Neural Networks by Sparse Kernel Selection

Binary Neural Network (BNN) represents convolution weights with 1-bit values, which enhances the efficiency of storage and computation. This paper is motivated by a previously revealed phenomenon that the binary kernels in successful BNNs are nearly power-law distributed: their values are mostly clustered into a small number of codewords. This phenomenon encourages us to compact typical BNNs and obtain further close performance through learning non-repetitive kernels within a binary kernel subspace. Specifically, we regard the binarization process as kernel grouping in terms of a binary codebook, and our task lies in learning to select a smaller subset of codewords from the full codebook. We then leverage the Gumbel-Sinkhorn technique to approximate the codeword selection process, and develop the Permutation Straight-Through Estimator (PSTE) that is able to not only optimize the selection process end-to-end but also maintain the non-repetitive occupancy of selected codewords. Experiments verify that our method reduces both the model size and bit-wise computational costs, and achieves accuracy improvements compared with state-of-the-art BNNs under comparable budgets.

cs.CV↗

arXiv2023

3D Human Pose Lifting with Grid Convolution

Existing lifting networks for regressing 3D human poses from 2D single-view poses are typically constructed with linear layers based on graph-structured representation learning. In sharp contrast to them, this paper presents Grid Convolution (GridConv), mimicking the wisdom of regular convolution operations in image space. GridConv is based on a novel Semantic Grid Transformation (SGT) which leverages a binary assignment matrix to map the irregular graph-structured human pose onto a regular weave-like grid pose representation joint by joint, enabling layer-wise feature learning with GridConv operations. We provide two ways to implement SGT, including handcrafted and learnable designs. Surprisingly, both designs turn out to achieve promising results and the learnable one is better, demonstrating the great potential of this new lifting representation learning formulation. To improve the ability of GridConv to encode contextual cues, we introduce an attention module over the convolutional kernel, making grid convolution operations input-dependent, spatial-aware and grid-specific. We show that our fully convolutional grid lifting network outperforms state-of-the-art methods with noticeable margins under (1) conventional evaluation on Human3.6M and (2) cross-evaluation on MPI-INF-3DHP. Code is available at https://github.com/OSVAI/GridConv

cs.CV↗

arXiv2022

Omni-Dimensional Dynamic Convolution

Learning a single static convolutional kernel in each convolutional layer is the common training paradigm of modern Convolutional Neural Networks (CNNs). Instead, recent research in dynamic convolution shows that learning a linear combination of $n$ convolutional kernels weighted with their input-dependent attentions can significantly improve the accuracy of light-weight CNNs, while maintaining efficient inference. However, we observe that existing works endow convolutional kernels with the dynamic property through one dimension (regarding the convolutional kernel number) of the kernel space, but the other three dimensions (regarding the spatial size, the input channel number and the output channel number for each convolutional kernel) are overlooked. Inspired by this, we present Omni-dimensional Dynamic Convolution (ODConv), a more generalized yet elegant dynamic convolution design, to advance this line of research. ODConv leverages a novel multi-dimensional attention mechanism with a parallel strategy to learn complementary attentions for convolutional kernels along all four dimensions of the kernel space at any convolutional layer. As a drop-in replacement of regular convolutions, ODConv can be plugged into many CNN architectures. Extensive experiments on the ImageNet and MS-COCO datasets show that ODConv brings solid accuracy boosts for various prevailing CNN backbones including both light-weight and large ones, e.g., 3.77%~5.71%|1.86%~3.72% absolute top-1 improvements to MobivleNetV2|ResNet family on the ImageNet dataset. Intriguingly, thanks to its improved feature learning ability, ODConv with even one single kernel can compete with or outperform existing dynamic convolution counterparts with multiple kernels, substantially reducing extra parameters. Furthermore, ODConv is also superior to other attention modules for modulating the output features or the convolutional weights.

cs.CV↗

arXiv2022

Efficient Meta-Tuning for Content-aware Neural Video Delivery

Recently, Deep Neural Networks (DNNs) are utilized to reduce the bandwidth and improve the quality of Internet video delivery. Existing methods train corresponding content-aware super-resolution (SR) model for each video chunk on the server, and stream low-resolution (LR) video chunks along with SR models to the client. Although they achieve promising results, the huge computational cost of network training limits their practical applications. In this paper, we present a method named Efficient Meta-Tuning (EMT) to reduce the computational cost. Instead of training from scratch, EMT adapts a meta-learned model to the first chunk of the input video. As for the following chunks, it fine-tunes the partial parameters selected by gradient masking of previous adapted model. In order to achieve further speedup for EMT, we propose a novel sampling strategy to extract the most challenging patches from video frames. The proposed strategy is highly efficient and brings negligible additional cost. Our method significantly reduces the computational cost and achieves even better performance, paving the way for applying neural video delivery techniques to practical applications. We conduct extensive experiments based on various efficient SR architectures, including ESPCN, SRCNN, FSRCNN and EDSR-1, demonstrating the generalization ability of our work. The code is released at \url{https://github.com/Neural-video-delivery/EMT-Pytorch-ECCV2022}.

cs.CV↗

arXiv2021

Sub-bit Neural Networks: Learning to Compress and Accelerate Binary Neural Networks

In the low-bit quantization field, training Binary Neural Networks (BNNs) is the extreme solution to ease the deployment of deep models on resource-constrained devices, having the lowest storage cost and significantly cheaper bit-wise operations compared to 32-bit floating-point counterparts. In this paper, we introduce Sub-bit Neural Networks (SNNs), a new type of binary quantization design tailored to compress and accelerate BNNs. SNNs are inspired by an empirical observation, showing that binary kernels learnt at convolutional layers of a BNN model are likely to be distributed over kernel subsets. As a result, unlike existing methods that binarize weights one by one, SNNs are trained with a kernel-aware optimization framework, which exploits binary quantization in the fine-grained convolutional kernel space. Specifically, our method includes a random sampling step generating layer-specific subsets of the kernel space, and a refinement step learning to adjust these subsets of binary kernels via optimization. Experiments on visual recognition benchmarks and the hardware deployment on FPGA validate the great potentials of SNNs. For instance, on ImageNet, SNNs of ResNet-18/ResNet-34 with 0.56-bit weights achieve 3.13/3.33 times runtime speed-up and 1.8 times compression over conventional BNNs with moderate drops in recognition accuracy. Promising results are also obtained when applying SNNs to binarize both weights and activations. Our code is available at https://github.com/yikaiw/SNN.

cs.CV↗