Rapid development of big data and high-performance computing have encouraged explosive studies of deep learning in geoscience. However, most studies only take single-type data as input, frittering away invaluable multisource, multi-scale information. We develop a general architecture of hybrid deep neural networks (HDNNs) to support mixed inputs. Regarding as a combination of feature learning and target learning, the new proposed networks provide great capacity in high-hierarchy feature extraction and in-depth data mining. Furthermore, the hybrid architecture is an aggregation of multiple networks, demonstrating good flexibility and wide applicability. The configuration of multiple networks depends on application tasks and varies with inputs and targets. Concentrating on reservoir production prediction, a specific HDNN model is configured and applied to an oil development block. Considering their contributions to hydrocarbon production, core photos, logging images and curves, geologic and engineering parameters can all be taken as inputs. After preprocessing, the mixed inputs are prepared as regular-sampled structural and numerical data. For feature learning, convolutional neural networks (CNN) and multilayer perceptron (MLP) network are configured to separately process structural and numerical inputs. Learned features are then concatenated and fed to subsequent networks for target learning. Comparison with typical MLP model and CNN model highlights the superiority of proposed HDNN model with high accuracy and good generalization.
Georgia Institute of Technology（佐治亚理工学院）；Nanjing University（南京大学）；Rice University（莱斯大学）
Multiplication is arguably the most cost-dominant operation in modern deep neural networks (DNNs), limiting their achievable efficiency and thus more extensive deployment in resource-constrained applications. To tackle this limitation, pioneering works have developed handcrafted multiplication-free DNNs, which require expert knowledge and time-consuming manual iteration, calling for fast development tools. To this end, we propose a Neural Architecture Search and Acceleration framework dubbed NASA, which enables automated multiplication-reduced DNN development and integrates a dedicated multipl…
Rolling bearings are one of the most widely used bearings in industrial machines. Deterioration in the condition of rolling bearings can result in the total failure of rotating machinery. AI-based methods are widely applied in the diagnosis of rolling bearings. Hybrid NN-based methods have been shown to achieve the best diagnosis results. Typically, raw data is generated from accelerometers mounted on the machine housing. However, the diagnostic utility of each signal is highly dependent on the location of the corresponding accelerometer. This paper proposes a novel hybrid CNN-MLP model-based …
We present a novel technique, called Term Revealing (TR), for furthering quantization at run time for improved performance of Deep Neural Networks (DNNs) already quantized with conventional quantization methods. TR operates on power-of-two terms in binary expressions of values. In computing a dot-product computation, TR dynamically selects a fixed number of largest terms to use from the values of the two vectors in the dot product. By exploiting normal-like weight and data distributions typically present in DNNs, TR has a minimal impact on DNN model performance (i.e., accuracy or perplexity). …
It is promising to solve linear inverse problems by unfolding iterative algorithms (e.g., iterative shrinkage thresholding algorithm (ISTA)) as deep neural networks (DNNs) with learnable parameters. However, existing ISTA-based unfolded algorithms restrict the network architectures for iterative updates with the partial weight coupling structure to guarantee convergence. In this paper, we propose hybrid ISTA to unfold ISTA with both pre-computed and learned parameters by incorporating free-form DNNs (i.e., DNNs with arbitrary feasible and reasonable network architectures), while ensuring theor…
Large DNNs with mixed-precision quantization can achieve ultra-high compression while retaining high classification performance. However, because of the challenges in finding an accurate metric that can guide the optimization process, these methods either sacrifice significant performance compared to the 32-bit floating-point (FP-32) baseline or rely on a compute-expensive, iterative training policy that requires the availability of a pre-trained baseline. To address this issue, this paper presents BMPQ, a training method that uses bit gradients to analyze layer sensitivities and yield mixed-p…
KTH Royal Institute of Technology（瑞典皇家理工学院）；Imperial College London（帝国理工学院）
In this work, we develop a novel input feature selection framework for ReLU-based deep neural networks (DNNs), which builds upon a mixed-integer optimization approach. While the method is generally applicable to various classification tasks, we focus on finding input features for image classification for clarity of presentation. The idea is to use a trained DNN, or an ensemble of trained DNNs, to identify the salient input features. The input feature selection is formulated as a sequence of mixed-integer linear programming (MILP) problems that find sets of sparse inputs that maximize the class…
Modern approaches to supervised learning like deep neural networks (DNNs) typically implicitly assume that observed responses are statistically independent. In contrast, correlated data are prevalent in real-life large-scale applications, with typical sources of correlation including spatial, temporal and clustering structures. These correlations are either ignored by DNNs, or ad-hoc solutions are developed for specific use cases. We propose to use the mixed models framework to handle correlated data in DNNs. By treating the effects underlying the correlation structure as random effects, mixed…
Hardware-Aware DNN Compression via Diverse Pruning and Mixed-Precision Quantization
Aristotle University of Thessaloniki（亚里士多德大学）；Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）；Southern Illinois University Carbondale（南伊利诺伊大学卡本代尔分校）；University of Patras（帕特雷大学）
With a growing need to enable intelligence in embedded devices in the Internet of Things (IoT) era, secure hardware implementation of Deep Neural Networks (DNNs) has become imperative. We will focus on how to address adversarial robustness for DNNs through efficiency-driven hardware optimizations. Since memory (specifically, dot-product operations) is a key energy-spending component for DNNs, hardware approaches in the past have focused on optimizing the memory. One such approach is approximate digital CMOS memories with hybrid 6T-8T SRAM cells that enable supply voltage (Vdd) scaling yielding…
Adversarial examples, crafted by adding imperceptible perturbations to natural inputs, can easily fool deep neural networks (DNNs). One of the most successful methods for training adversarially robust DNNs is solving a nonconvex-nonconcave minimax problem with an adversarial training (AT) algorithm. However, among the many AT algorithms, only Dynamic AT (DAT) and You Only Propagate Once (YOPO) guarantee convergence to a stationary point. In this work, we generalize the stochastic primal-dual hybrid gradient algorithm to develop semi-implicit hybrid gradient methods (SI-HGs) for finding station…
Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences（中国科学院大学杭州高等研究院）；State key laboratory of Modern Optical Instrumentation, Department of Optical Engineering, Zhejiang University（浙江大学光电工程系现代光学仪器国家重点实验室）；Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences（中国科学院上海光学精密机械研究所）；Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences（中国科学院大学材料科学与光电子工程中心）
Deep Neural Networks (DNNs) have been shown to be vulnerable to adversarial attacks, wherein, a model gets fooled by applying slight perturbations on the input. With the advent of Internet-of-Things and the necessity to enable intelligence in embedded devices, low-power and secure hardware implementation of DNNs is vital. In this paper, we investigate the use of quantization to potentially resist adversarial attacks. Several recent studies have reported remarkable results in reducing the energy requirement of a DNN through quantization. However, no prior work has considered the relationship be…
This paper presents a dynamic network rewiring (DNR) method to generate pruned deep neural network (DNN) models that are robust against adversarial attacks yet maintain high accuracy on clean images. In particular, the disclosed DNR method is based on a unified constrained optimization formulation using a hybrid loss function that merges ultra-high model compression with robust adversarial training. This training strategy dynamically adjusts inter-layer connectivity based on per-layer normalized momentum computed from the hybrid loss function. In contrast to existing robust pruning frameworks …
The performance of deep neural networks crucially depends on good hyperparameter configurations. Bayesian optimization is a powerful framework for optimizing the hyperparameters of DNNs. These methods need sufficient evaluation data to approximate and minimize the validation error function of hyperparameters. However, the expensive evaluation cost of DNNs leads to very few evaluation data within a limited time, which greatly reduces the efficiency of Bayesian optimization. Besides, the previous researches focus on using the complete evaluation data to conduct Bayesian optimization, and ignore …
Xi’an Jiaotong University（西安交通大学）；The Hong Kong Polytechnic University（香港理工大学）
Mixed-precision quantization mostly predetermines the model bit-width settings before actual training due to the non-differential bit-width sampling process, obtaining sub-optimal performance. Worse still, the conventional static quality-consistent training setting, i.e., all data is assumed to be of the same quality across training and inference, overlooks data quality changes in real-world applications which may lead to poor robustness of the quantized models. In this paper, we propose a novel Data Quality-aware Mixed-precision Quantization framework, dubbed DQMQ, to dynamically adapt quanti…
King’s College London（伦敦国王学院）；Arizona State University（亚利桑那州立大学）
The cost involved in training deep neural networks (DNNs) on von-Neumann architectures has motivated the development of novel solutions for efficient DNN training accelerators. We propose a hybrid in-memory computing (HIC) architecture for the training of DNNs on hardware accelerators that results in memory-efficient inference and outperforms baseline software accuracy in benchmark tasks. We introduce a weight representation technique that exploits both binary and multi-level phase-change memory (PCM) devices, and this leads to a memory-efficient inference accelerator. Unlike previous in-memor…
This work proposes a novel Energy-Aware Network Operator Search (ENOS) approach to address the energy-accuracy trade-offs of a deep neural network (DNN) accelerator. In recent years, novel inference operators have been proposed to improve the computational efficiency of a DNN. Augmenting the operators, their corresponding novel computing modes have also been explored. However, simplification of DNN operators invariably comes at the cost of lower accuracy, especially on complex processing tasks. Our proposed ENOS framework allows an optimal layer-wise integration of inference operators and comp…
Development of Hybrid Artificial Intelligence Training on Real and Synthetic Data: Benchmark on Two Mixed Training Strategies
Faculty of Engineering and Computer Science（工程与计算机科学学院）；Osnabrück University of Applied Sciences（奥斯纳布鲁克应用科学大学）；Institute of Cognitive Science（认知科学研究所）；Osnabrück University（奥斯纳布鲁克大学）
