In this paper, the feasibility of adopting an intelligent reflective surface (IRS) in a cell-free wireless communication system is studied. The received signal-to-noise ratio (SNR) for this IRS-enabled cell-free set-up is optimized by adjusting phase-shifts of the passive reflective elements. Then, tight approximations for the probability density function and the cumulative distribution function for this optimal SNR are derived for Rayleigh fading. To investigate the performance of this system model, tight bounds/approximations for the achievable rate and outage probability are derived in closed form. The impact of discrete phase-shifts is modeled, and the corresponding detrimental effects are investigated by deriving an upper bound for the achievable rate in the presence of phase-shift quantization errors. Monte-Carlo simulations are used to validate our statistical characterization of the optimal SNR, and the corresponding analysis is used to investigate the performance gains of the proposed system model. We reveal that IRS-assisted communications can boost the performance of cell-free wireless architectures.
Transformers' quadratic complexity with respect to the input sequence length has motivated a body of work on efficient sparse approximations to softmax. An alternative path, used by entmax transformers, consists of having built-in exact sparse attention; however this approach still requires quadratic computation. In this paper, we propose Sparsefinder, a simple model trained to identify the sparsity pattern of entmax attention before computing it. We experiment with three variants of our method, based on distances, quantization, and clustering, on two tasks: machine translation (attention in the decoder) and masked language modeling (encoder-only). Our work provides a new angle to study model efficiency by doing extensive analysis of the tradeoff between the sparsity and recall of the predicted attention graph. This allows for detailed comparison between different models along their Pareto curves, important to guide future benchmarks for sparse attention models.
Transformer-based architectures have become the de-facto standard models for a wide range of Natural Language Processing tasks. However, their memory footprint and high latency are prohibitive for efficient deployment and inference on resource-limited devices. In this work, we explore quantization for transformers. We show that transformers have unique quantization challenges -- namely, high dynamic activation ranges that are difficult to represent with a low bit fixed-point format. We establish that these activations contain structured outliers in the residual connections that encourage specific attention patterns, such as attending to the special separator token. To combat these challenges, we present three solutions based on post-training quantization and quantization-aware training, each with a different set of compromises for accuracy, model size, and ease of use. In particular, we introduce a novel quantization scheme -- per-embedding-group quantization. We demonstrate the effectiveness of our methods on the GLUE benchmark using BERT, establishing state-of-the-art results for post-training quantization. Finally, we show that transformer weights and embeddings can be quantized to ultra-low bit-widths, leading to significant memory savings with a minimum accuracy loss. Our source code is available at~\url{https://github.com/qualcomm-ai-research/transformer-quantization}.
Classic and deep generalized canonical correlation analysis (GCCA) algorithms seek low-dimensional common representations of data entities from multiple ``views'' (e.g., audio and image) using linear transformations and neural networks, respectively. When the views are acquired and stored at different computing agents (e.g., organizations and edge devices) and data sharing is undesired due to privacy or communication cost considerations, federated learning-based GCCA is well-motivated. In federated learning, the views are kept locally at the agents and only derived, limited information exchange with a central server is allowed. However, applying existing GCCA algorithms onto such federated learning settings may incur prohibitively high communication overhead. This work puts forth a communication-efficient federated learning framework for both linear and deep GCCA under the maximum variance (MAX-VAR) formulation. The overhead issue is addressed by aggressively compressing (via quantization) the exchanging information between the computing agents and a central controller. Compared to the unquantized version, our empirical study shows that the proposed algorithm enjoys a substantial reduction of communication overheads with virtually no loss in accuracy and convergence speed. Rigorous convergence analyses are also presented, which is a nontrivial effort. Generic federated optimization results do not cover the special problem structure of GCCA. Our result shows that the proposed algorithms for both linear and deep GCCA converge to critical points at a sublinear rate, even under heavy quantization and stochastic approximations. In addition, in the linear MAX-VAR case, the quantized algorithm approaches a global optimum in a geometric rate under reasonable conditions. Synthetic and real-data experiments are used to showcase the effectiveness of the proposed approach.
Deep learning has shown a tremendous growth in hashing techniques for image retrieval. Recently, Transformer has emerged as a new architecture by utilizing self-attention without convolution. Transformer is also extended to Vision Transformer (ViT) for the visual recognition with a promising performance on ImageNet. In this paper, we propose a Vision Transformer based Hashing (VTS) for image retrieval. We utilize the pre-trained ViT on ImageNet as the backbone network and add the hashing head. The proposed VTS model is fine tuned for hashing under six different image retrieval frameworks, including Deep Supervised Hashing (DSH), HashNet, GreedyHash, Improved Deep Hashing Network (IDHN), Deep Polarized Network (DPN) and Central Similarity Quantization (CSQ) with their objective functions. We perform the extensive experiments on CIFAR10, ImageNet, NUS-Wide, and COCO datasets. The proposed VTS based image retrieval outperforms the recent state-of-the-art hashing techniques with a great margin. We also find the proposed VTS model as the backbone network is better than the existing networks, such as AlexNet and ResNet. The code is released at \url{https://github.com/shivram1987/VisionTransformerHashing}.
Model binarization is an effective method of compressing neural networks and accelerating their inference process. However, a significant performance gap still exists between the 1-bit model and the 32-bit one. The empirical study shows that binarization causes a great loss of information in the forward and backward propagation. We present a novel Distribution-sensitive Information Retention Network (DIR-Net) that retains the information in the forward and backward propagation by improving internal propagation and introducing external representations. The DIR-Net mainly relies on three technical contributions: (1) Information Maximized Binarization (IMB): minimizing the information loss and the binarization error of weights/activations simultaneously by weight balance and standardization; (2) Distribution-sensitive Two-stage Estimator (DTE): retaining the information of gradients by distribution-sensitive soft approximation by jointly considering the updating capability and accurate gradient; (3) Representation-align Binarization-aware Distillation (RBD): retaining the representation information by distilling the representations between full-precision and binarized networks. The DIR-Net investigates both forward and backward processes of BNNs from the unified information perspective, thereby providing new insight into the mechanism of network binarization. The three techniques in our DIR-Net are versatile and effective and can be applied in various structures to improve BNNs. Comprehensive experiments on the image classification and objective detection tasks show that our DIR-Net consistently outperforms the state-of-the-art binarization approaches under mainstream and compact architectures, such as ResNet, VGG, EfficientNet, DARTS, and MobileNet. Additionally, we conduct our DIR-Net on real-world resource-limited devices which achieves 11.1x storage saving and 5.4x speedup.
Massive amounts of data have led to the training of large-scale machine learning models on a single worker inefficient. Distributed machine learning methods such as Parallel-SGD have received significant interest as a solution to tackle this problem. However, the performance of distributed systems does not scale linearly with the number of workers due to the high network communication cost for synchronizing gradients and parameters. Researchers have proposed techniques such as quantization and sparsification to alleviate this problem by compressing the gradients. Most of the compression schemes result in compressed gradients that cannot be directly aggregated with efficient protocols such as all-reduce. In this paper, we present a set of all-reduce compatible gradient compression schemes which significantly reduce the communication overhead while maintaining the performance of vanilla SGD. We present the results of our experiments with the CIFAR10 dataset and observations derived during the process. Our compression methods perform better than the in-built methods currently offered by the deep learning frameworks. Code is available at the repository: \url{https://github.com/vineeths96/Gradient-Compression}.
With the recent increase in the computational power of modern mobile devices, machine learning-based heavy tasks such as face detection and speech recognition are now integral parts of such devices. This requires frameworks to execute machine learning models (e.g., Deep Neural Networks) on mobile devices. Although there exist studies on the accuracy and performance of these frameworks, the quality of on-device deep learning frameworks, in terms of their robustness, has not been systematically studied yet. In this paper, we empirically compare two on-device deep learning frameworks with three adversarial attacks on three different model architectures. We also use both the quantized and unquantized variants for each architecture. The results show that, in general, neither of the deep learning frameworks is better than the other in terms of robustness, and there is not a significant difference between the PC and mobile frameworks either. However, in cases like Boundary attack, mobile version is more robust than PC. In addition, quantization improves robustness in all cases when moving from PC to mobile.
Recurrent neural networks (RNN) are used in many real-world text and speech applications. They include complex modules such as recurrence, exponential-based activation, gate interaction, unfoldable normalization, bi-directional dependence, and attention. The interaction between these elements prevents running them on integer-only operations without a significant performance drop. Deploying RNNs that include layer normalization and attention on integer-only arithmetic is still an open problem. We present a quantization-aware training method for obtaining a highly accurate integer-only recurrent neural network (iRNN). Our approach supports layer normalization, attention, and an adaptive piecewise linear approximation of activations (PWL), to serve a wide range of RNNs on various applications. The proposed method is proven to work on RNN-based language models and challenging automatic speech recognition, enabling AI applications on the edge. Our iRNN maintains similar performance as its full-precision counterpart, their deployment on smartphones improves the runtime performance by $2\times$, and reduces the model size by $4\times$.
The security and privacy concerns along with the amount of data that is required to be processed on regular basis has pushed processing to the edge of the computing systems. Deploying advanced Neural Networks (NN), such as deep neural networks (DNNs) and spiking neural networks (SNNs), that offer state-of-the-art results on resource-constrained edge devices is challenging due to the stringent memory and power/energy constraints. Moreover, these systems are required to maintain correct functionality under diverse security and reliability threats. This paper first discusses existing approaches to address energy efficiency, reliability, and security issues at different system layers, i.e., hardware (HW) and software (SW). Afterward, we discuss how to further improve the performance (latency) and the energy efficiency of Edge AI systems through HW/SW-level optimizations, such as pruning, quantization, and approximation. To address reliability threats (like permanent and transient faults), we highlight cost-effective mitigation techniques, like fault-aware training and mapping. Moreover, we briefly discuss effective detection and protection techniques to address security threats (like model and data corruption). Towards the end, we discuss how these techniques can be combined in an integrated cross-layer framework for realizing robust and energy-efficient Edge AI systems.
This paper focuses on channel estimation in single-user and multi-user MIMO systems with multi-antenna base stations equipped with 1-bit spatial sigma-delta analog-to-digital converters (ADCs). A careful selection of the quantization voltage level and phase shift used in the feedback loop of 1-bit sigma-delta ADCs is critical to improve its effective resolution. We first develop a quantization noise model for 1-bit spatial sigma-delta ADCs. Using the developed noise model, we then present a two-step channel estimation algorithm to estimate a multipath channel parameterized by the gains, angles of arrival (AoAs), and angles of departure (AoDs). Specifically, in the first step, the AoAs and path gains are estimated using uplink pilots, which excite all the angles uniformly. Next, in the second step, the AoDs are estimated by progressively refining uplink beams through a recursive bisection procedure. For this algorithm, we propose a technique to select the quantization voltage level and phase shift. Through numerical simulations, we demonstrate that with the proposed parametric channel estimation algorithm, MIMO systems with 1-bit spatial sigma-delta ADCs perform significantly better than those with regular 1-bit ADCs and are on par with MIMO systems with high-resolution ADCs.
Neural network quantization enables the deployment of models on edge devices. An essential requirement for their hardware efficiency is that the quantizers are hardware-friendly: uniform, symmetric, and with power-of-two thresholds. To the best of our knowledge, current post-training quantization methods do not support all of these constraints simultaneously. In this work, we introduce a hardware-friendly post training quantization (HPTQ) framework, which addresses this problem by synergistically combining several known quantization methods. We perform a large-scale study on four tasks: classification, object detection, semantic segmentation and pose estimation over a wide variety of network architectures. Our extensive experiments show that competitive results can be obtained under hardware-friendly constraints.
To bridge the ever increasing gap between deep neural networks' complexity and hardware capability, network quantization has attracted more and more research attention. The latest trend of mixed precision quantization takes advantage of hardware's multiple bit-width arithmetic operations to unleash the full potential of network quantization. However, this also results in a difficult integer programming formulation, and forces most existing approaches to use an extremely time-consuming search process even with various relaxations. Instead of solving a problem of the original integer programming, we propose to optimize a proxy metric, the concept of network orthogonality, which is highly correlated with the loss of the integer programming but also easy to optimize with linear programming. This approach reduces the search time and required data amount by orders of magnitude, with little compromise on quantization accuracy. Specifically, we achieve 72.08% Top-1 accuracy on ResNet-18 with 6.7Mb, which does not require any searching iterations. Given the high efficiency and low data dependency of our algorithm, we used it for the post-training quantization, which achieve 71.27% Top-1 accuracy on MobileNetV2 with only 1.5Mb. Our code is available at https://github.com/MAC-AutoML/OMPQ.
Dense retrieval methods have shown great promise over sparse retrieval methods in a range of NLP problems. Among them, dense phrase retrieval-the most fine-grained retrieval unit-is appealing because phrases can be directly used as the output for question answering and slot filling tasks. In this work, we follow the intuition that retrieving phrases naturally entails retrieving larger text blocks and study whether phrase retrieval can serve as the basis for coarse-level retrieval including passages and documents. We first observe that a dense phrase-retrieval system, without any retraining, already achieves better passage retrieval accuracy (+3-5% in top-5 accuracy) compared to passage retrievers, which also helps achieve superior end-to-end QA performance with fewer passages. Then, we provide an interpretation for why phrase-level supervision helps learn better fine-grained entailment compared to passage-level supervision, and also show that phrase retrieval can be improved to achieve competitive performance in document-retrieval tasks such as entity linking and knowledge-grounded dialogue. Finally, we demonstrate how phrase filtering and vector quantization can reduce the size of our index by 4-10x, making dense phrase retrieval a practical and versatile solution in multi-granularity retrieval.
The deployment of artificial neural networks-based optical channel equalizers on edge-computing devices is critically important for the next generation of optical communication systems. However, this is still a highly challenging problem, mainly due to the computational complexity of the artificial neural networks (NNs) required for the efficient equalization of nonlinear optical channels with large dispersion-induced memory. To implement the NN-based optical channel equalizer in hardware, a substantial complexity reduction is needed, while we have to keep an acceptable performance level of the simplified NN model. In this work, we address the complexity reduction problem by applying pruning and quantization techniques to an NN-based optical channel equalizer. We use an exemplary NN architecture, the multi-layer perceptron (MLP), to mitigate the impairments for 30GBd 1000km transmission over a standard single-mode fiber, and demonstrate that it is feasible to reduce the equalizer's memory by up to 87.12%, and its complexity by up to 78.34%, without noticeable performance degradation. In addition to this, we accurately define the computational complexity of a compressed NN-based equalizer in the digital signal processing (DSP) sense. Further, we examine the impact of using different CPU and GPU settings on the power consumption and latency for the compressed equalizer. We also verify the developed technique experimentally, by implementing the reduced NN equalizer on two standard edge-computing hardware units: Raspberry Pi 4 and Nvidia Jetson Nano, which are used to process the data generated via simulating the signal's propagation down the optical-fiber system.
The recent breakthroughs of deep neural networks (DNNs) and the advent of billions of Internet of Things (IoT) devices have excited an explosive demand for intelligent IoT devices equipped with domain-specific DNN accelerators. However, the deployment of DNN accelerator enabled intelligent functionality into real-world IoT devices still remains particularly challenging. First, powerful DNNs often come at prohibitive complexities, whereas IoT devices often suffer from stringent resource constraints. Second, while DNNs are vulnerable to adversarial attacks especially on IoT devices exposed to complex real-world environments, many IoT applications require strict security. Existing DNN accelerators mostly tackle only one of the two aforementioned challenges (i.e., efficiency or adversarial robustness) while neglecting or even sacrificing the other. To this end, we propose a 2-in-1 Accelerator, an integrated algorithm-accelerator co-design framework aiming at winning both the adversarial robustness and efficiency of DNN accelerators. Specifically, we first propose a Random Precision Switch (RPS) algorithm that can effectively defend DNNs against adversarial attacks by enabling random DNN quantization as an in-situ model switch. Furthermore, we propose a new precision-scalable accelerator featuring (1) a new precision-scalable MAC unit architecture which spatially tiles the temporal MAC units to boost both the achievable efficiency and flexibility and (2) a systematically optimized dataflow that is searched by our generic accelerator optimizer. Extensive experiments and ablation studies validate that our 2-in-1 Accelerator can not only aggressively boost both the adversarial robustness and efficiency of DNN accelerators under various attacks, but also naturally support instantaneous robustness-efficiency trade-offs adapting to varied resources without the necessity of DNN retraining.
Deep hashing approaches, including deep quantization and deep binary hashing, have become a common solution to large-scale image retrieval due to their high computation and storage efficiency. Most existing hashing methods cannot produce satisfactory results for fine-grained retrieval, because they usually adopt the outputs of the last CNN layer to generate binary codes. Since deeper layers tend to summarize visual clues, e.g., texture, into abstract semantics, e.g., dogs and cats, the feature produced by the last CNN layer is less effective in capturing subtle but discriminative visual details that mostly exist in shallow layers. To improve fine-grained image hashing, we propose Pyramid Hybrid Pooling Quantization (PHPQ). Specifically, we propose a Pyramid Hybrid Pooling (PHP) module to capture and preserve fine-grained semantic information from multi-level features, which emphasizes the subtle discrimination of different sub-categories. Besides, we propose a learnable quantization module with a partial codebook attention mechanism, which helps to optimize the most relevant codewords and improves the quantization. Comprehensive experiments on two widely-used public benchmarks, i.e., CUB-200-2011 and Stanford Dogs, demonstrate that PHPQ outperforms state-of-the-art methods.
A general nonlinear $1$st-order consensus-based solution for distributed constrained convex optimization is proposed with network resource allocation applications. The solution is used to optimize continuously-differentiable strictly convex cost functions over weakly-connected undirected networks, while it is anytime feasible and models various nonlinearities to account for imperfections and constraints on the (physical model of) agents in terms of limited actuation capabilities, e.g., quantization and saturation. Due to such inherent nonlinearities, the existing linear solutions considering ideal agent models may not necessarily converge with guaranteed optimality and anytime feasibility. Some applications also impose specific nonlinearities, e.g., convergence in fixed/finite-time or sign-based robust disturbance-tolerant dynamics. Our proposed distributed protocol generalizes such nonlinear models. Putting convex set analysis together with nonsmooth Lyapunov analysis, we prove convergence, (i) regardless of the particular type of nonlinearity, and (ii) with weak network-connectivity requirements (uniform-connectivity).
We characterize the frequency response of channel-interleaved photonic analog-to-digital converters (CI-PADCs) theoretically and experimentally. The CI-PADC is composed of a photonic frontend for photonic sampling and an electronic backend for quantization. The photonic frontend includes a photonic sampling pulse generator for directly high-speed sampling and an optical time-division demultiplexer (OTDM) for channel demultiplexing. It is found that the frequency response of the CI-PADC is influenced by both the photonic sampling pulses and the OTDM, of which the combined impact can be characterized through demultiplexed pulse trains. First, the frequency response can be divided into multiple frequency intervals and the range of the frequency interval equals the repetition rate of demultiplexed pulse trains. Second, the analog bandwidth of the CI-PADC is determined by the optical spectral bandwidth of demultiplexed pulse trains which is broadened in the OTDM. Further, the effect of the OTDM is essential for enlarging the analog bandwidth of the CI-PADC employing the photonic sampling pulses with a limited optical spectral bandwidth.
Improving Transformer efficiency has become increasingly attractive recently. A wide range of methods has been proposed, e.g., pruning, quantization, new architectures and etc. But these methods are either sophisticated in implementation or dependent on hardware. In this paper, we show that the efficiency of Transformer can be improved by combining some simple and hardware-agnostic methods, including tuning hyper-parameters, better design choices and training strategies. On the WMT news translation tasks, we improve the inference efficiency of a strong Transformer system by 3.80X on CPU and 2.52X on GPU. The code is publicly available at https://github.com/Lollipop321/mini-decoder-network.