An-Yeu

Cheng-Yen Hsieh, Yu-Chuan Chuang, An-Yeu et al.

Most existing studies improve the efficiency of Split learning (SL) by compressing the transmitted features. However, most works focus on dimension-wise compression that transforms high-dimensional features into a low-dimensional space. In this paper, we propose circular convolution-based batch-wise compression for SL (C3-SL) to compress multiple features into one single feature. To avoid information loss while merging multiple features, we exploit the quasi-orthogonality of features in high-dimensional space with circular convolution and superposition. To the best of our knowledge, we are the first to explore the potential of batch-wise compression under the SL scenario. Based on the simulation results on CIFAR-10 and CIFAR-100, our method achieves a 16x compression ratio with negligible accuracy drops compared with the vanilla SL. Moreover, C3-SL significantly reduces 1152x memory and 2.25x computation overhead compared to the state-of-the-art dimension-wise compression method.

Hao-Wei Chiang, Chi-Tse Huang, Hsiang-Yun Cheng et al.

While memory-augmented neural networks (MANNs) offer an effective solution for few-shot learning (FSL) by integrating deep neural networks with external memory, the capacity requirements and energy overhead of data movement become enormous due to the large number of support vectors in many-class FSL scenarios. Various in-memory search solutions have emerged to improve the energy efficiency of MANNs. NAND-based multi-bit content addressable memory (MCAM) is a promising option due to its high density and large capacity. Despite its potential, MCAM faces limitations such as a restricted number of word lines, limited quantization levels, and non-ideal effects like varying string currents and bottleneck effects, which lead to significant accuracy drops. To address these issues, we propose several innovative methods. First, the Multi-bit Thermometer Code (MTMC) leverages the extensive capacity of MCAM to enhance vector precision using cumulative encoding rules, thereby mitigating the bottleneck effect. Second, the Asymmetric vector similarity search (AVSS) reduces the precision of the query vector while maintaining that of the support vectors, thereby minimizing the search iterations and improving efficiency in many-class scenarios. Finally, the Hardware-Aware Training (HAT) method optimizes controller training by modeling the hardware characteristics of MCAM, thus enhancing the reliability of the system. Our integrated framework reduces search iterations by up to 32 times, and increases overall accuracy by 1.58% to 6.94%.

Yu-Shan Tai, An-Yeu, Wu

Recently, diffusion models (DMs) have made significant strides in high-quality image generation. However, the multi-step denoising process often results in considerable computational overhead, impeding deployment on resource-constrained edge devices. Existing methods mitigate this issue by compressing models and adjusting the time step sequence. However, they overlook input redundancy and require lengthy search times. In this paper, we propose Coarse-to-Fine Diffusion Models with Time Step Sequence Redistribution. Recognizing indistinguishable early-stage generated images, we introduce Coarse-to-Fine Denoising (C2F) to reduce computation during coarse feature generation. Furthermore, we design Time Step Sequence Redistribution (TRD) for efficient sampling trajectory adjustment, requiring less than 10 minutes for search. Experimental results demonstrate that the proposed methods achieve near-lossless performance with an 80% to 90% reduction in computation on CIFAR10 and LSUN-Church.

Jiing-Ping Wang, Ming-Guang Lin, An-Yeu et al.

With the rise of Transformer models in NLP and CV domain, Multi-Head Attention has been proven to be a game-changer. However, its expensive computation poses challenges to the model throughput and efficiency, especially for the long sequence tasks. Exploiting the sparsity in attention has been proven to be an effective way to reduce computation. Nevertheless, prior works do not consider the various distributions among different heads and lack a systematic method to determine the threshold. To address these challenges, we propose Low-Precision Approximate Attention with Head-wise Trainable Threshold for Efficient Transformer (LATTE). LATTE employs a headwise threshold-based filter with the low-precision dot product and computation reuse mechanism to reduce the computation of MHA. Moreover, the trainable threshold is introduced to provide a systematic method for adjusting the thresholds and enable end-to-end optimization. Experimental results indicate LATTE can smoothly adapt to both NLP and CV tasks, offering significant computation savings with only a minor compromise in performance. Also, the trainable threshold is shown to be essential for the leverage between the performance and the computation. As a result, LATTE filters up to 85.16% keys with only a 0.87% accuracy drop in the CV task and 89.91% keys with a 0.86 perplexity increase in the NLP task.

Yu-Shan Tai, An-Yeu, Wu

Recent advancements in diffusion models have demonstrated remarkable success in various image generation tasks. Building upon these achievements, diffusion models have also been effectively adapted to image restoration tasks, e.g., super-resolution and deblurring, aiming to recover high-quality images from degraded inputs. Although existing zero-shot approaches enable pretrained diffusion models to perform restoration tasks without additional fine-tuning, these methods often suffer from prolonged iteration times in the denoising process. To address this limitation, we propose a Quick Bypass Mechanism (QBM), a strategy that significantly accelerates the denoising process by initializing from an intermediate approximation, effectively bypassing early denoising steps. Furthermore, recognizing that approximation may introduce inconsistencies, we introduce a Revised Reverse Process (RRP), which adjusts the weighting of random noise to enhance the stochasticity and mitigate potential disharmony. We validate proposed methods on ImageNet-1K and CelebA-HQ across multiple image restoration tasks, e.g., super-resolution, deblurring, and compressed sensing. Our experimental results show that the proposed methods can effectively accelerate existing methods while maintaining original performance.

Yu-Shan Tai, An-Yeu, Wu

While vision transformers (ViTs) have shown great potential in computer vision tasks, their intense computation and memory requirements pose challenges for practical applications. Existing post-training quantization methods leverage value redistribution or specialized quantizers to address the non-normal distribution in ViTs. However, without considering the asymmetry in activations and relying on hand-crafted settings, these methods often struggle to maintain performance under low-bit quantization. To overcome these challenges, we introduce SmoothQuant with bias term (SQ-b) to alleviate the asymmetry issue and reduce the clamping loss. We also introduce optimal scaling factor ratio search (OPT-m) to determine quantization parameters by a data-dependent mechanism automatically. To further enhance the compressibility, we incorporate the above-mentioned techniques and propose a mixed-precision post-training quantization framework for vision transformers (MPTQ-ViT). We develop greedy mixed-precision quantization (Greedy MP) to allocate layer-wise bit-width considering both model performance and compressibility. Our experiments on ViT, DeiT, and Swin demonstrate significant accuracy improvements compared with SOTA on the ImageNet dataset. Specifically, our proposed methods achieve accuracy improvements ranging from 0.90% to 23.35% on 4-bit ViTs with single-precision and from 3.82% to 78.14% on 5-bit fully quantized ViTs with mixed-precision.

Yi-Ta Chen, Yu-Chuan Chuang, An-Yeu et al.

In this paper, we propose an AdaBoost-assisted extreme learning machine for efficient online sequential classification (AOS-ELM). In order to achieve better accuracy in online sequential learning scenarios, we utilize the cost-sensitive algorithm-AdaBoost, which diversifying the weak classifiers, and adding the forgetting mechanism, which stabilizing the performance during the training procedure. Hence, AOS-ELM adapts better to sequentially arrived data compared with other voting based methods. The experiment results show AOS-ELM can achieve 94.41% accuracy on MNIST dataset, which is the theoretical accuracy bound performed by an original batch learning algorithm, AdaBoost-ELM. Moreover, with the forgetting mechanism, the standard deviation of accuracy during the online sequential learning process is reduced to 8.26x.