Machine Intelligence Applications Research Articles

Experimental demonstration of magnetic tunnel junction-based computational random-access memory

The conventional computing paradigm struggles to fulfill the rapidly growing demands from emerging applications, especially those for machine intelligence because much of the power and energy is consumed by constant data transfers between logic and memory modules. A new paradigm, called “computational random-access memory (CRAM),” has emerged to address this fundamental limitation. CRAM performs logic operations directly using the memory cells themselves, without having the data ever leave the memory. The energy and performance benefits of CRAM for both conventional and emerging applications have been well established by prior numerical studies. However, there is a lack of experimental demonstration and study of CRAM to evaluate its computational accuracy, which is a realistic and application-critical metric for its technological feasibility and competitiveness. In this work, a CRAM array based on magnetic tunnel junctions (MTJs) is experimentally demonstrated. First, basic memory operations, as well as 2-, 3-, and 5-input logic operations, are studied. Then, a 1-bit full adder with two different designs is demonstrated. Based on the experimental results, a suite of models has been developed to characterize the accuracy of CRAM computation. Scalar addition, multiplication, and matrix multiplication, which are essential building blocks for many conventional and machine intelligence applications, are evaluated and show promising accuracy performance. With the confirmation of MTJ-based CRAM’s accuracy, there is a strong case that this technology will have a significant impact on power- and energy-demanding applications of machine intelligence.

FastESN: Fast Echo State Network.

Echo state networks (ESNs) are reservoir computing-based recurrent neural networks widely used in pattern analysis and machine intelligence applications. In order to achieve high accuracy with large model capacity, ESNs usually contain a large-sized internal layer (reservoir), making the evaluation process too slow for some applications. In this work, we speed up the evaluation of ESN by building a reduced network called the fast ESN (fastESN) and achieve an ESN evaluation complexity independent of the original ESN size for the first time. FastESN is generated using three techniques. First, the high-dimensional state of the original ESN is approximated by a low-dimensional state through proper orthogonal decomposition (POD)-based projection. Second, the activation function evaluation number is reduced through the discrete empirical interpolation method (DEIM). Third, we show the directly generated fastESN has instability problems and provide a stabilization scheme as a solution. Through experiments on four popular benchmarks, we show that fastESN is able to accelerate the sparse storage-based ESN evaluation with a high parameter compression ratio and a fast evaluation speed.

Trusting machine intelligence: artificial intelligence and human-autonomy teaming in military operations

ABSTRACT Continuous advances in artificial intelligence has enabled higher levels of autonomy in military systems. As the role of machine-intelligence expands, effective co-operation between humans and autonomous systems will become an increasingly relevant aspect of future military operations. Successful human-autonomy teaming (HAT) requires establishing appropriate levels of trust in machine-intelligence, which can vary according to the context in which HAT occurs. The expansive body of literature on trust and automation, combined with newer contributions focused on autonomy in military systems, forms the basis of this study. Various aspects of trust within three general categories of machine intelligence applications are examined. These include data integration and analysis, autonomous systems in all domains, and decision-support applications. The issues related to appropriately calibrating trust levels varies within each category, as do the consequences of poorly aligned trust and potential mitigation measures.

Attention U-Net Based on Bi-ConvLSTM and Its Optimization for Smart Healthcare

As an important part of cyber–physical–social intelligence, artificial intelligence (AI)-driven smart healthcare is committed to promoting the application of human–machine hybrid augmented intelligence in the medical field, including AI-assisted medical image analysis and lesion recognition. Among them, deep learning models represented by fully convolutional networks (FCNs) have achieved excellent performance in medical image segmentation. However, limited by the complex structure of segmentation networks and the inherently redundant characteristics of convolutional operation, the scale of these models is extremely large. To further promote the application of machine intelligence in the field of medical image analysis, we propose an attention U-Net based on Bi-ConvLSTM (AUBC-Net) for accurate segmentation of medical images in this article. Different from classical U-Net, the proposed model deals with the potential association between decoding features and encoding features by bidirectional convolution LSTM. Furthermore, for the inherent redundancy characteristics of FCNs, we propose a lightweight feature generation strategy and optimize the calculation process of Bi-ConvLSTM based on tensor multilinear algebra, which can greatly reduce the number of network parameters. In addition, we have conducted the image segmentation experiments on two benchmark medical datasets, and the experimental results demonstrate that the proposed model can not only achieve better performance than existing methods, but also effectively compress network parameters while ensuring performance, which greatly facilitates AI-driven smart medical applications.

Bioinformatics Approaches Applied to the Discovery of Antifungal Peptides

Antifungal peptides (AFPs) comprise a group of substances with a broad spectrum of activities and complex action mechanisms. They develop in nature via an evolutionary process resulting from the interactions between hosts and pathogens. The AFP database is experimentally verified and curated from research articles, patents, and public databases. In this review, we compile information about the primary databases and bioinformatics tools that have been used in the discovery of AFPs during the last 15 years. We focus on the classification and prediction of AFPs using different physicochemical properties, such as polarity, hydrophobicity, hydrophilicity, mass, acidic, basic, and isoelectric indices, and other structural properties. Another method for discovering AFPs is the implementation of a peptidomic approach and bioinformatics filtering, which gave rise to a new family of peptides that exhibit a broad spectrum of antimicrobial activity against Candida albicans with low hemolytic effects. The application of machine intelligence in the sphere of biological sciences has led to the development of automated tools. The progress made in this area has also paved the way for producing new drugs more quickly and effectively. However, we also identified that further advancements are still needed to complete the AFP libraries.

Accelerating the discovery of antifungal peptides using deep temporal convolutional networks.

The application of machine intelligence in biological sciences has led to the development of several automated tools, thus enabling rapid drug discovery. Adding to this development is the ongoing COVID-19 pandemic, due to which researchers working in the field of artificial intelligence have acquired an active interest in finding machine learning-guided solutions for diseases like mucormycosis, which has emerged as an important post-COVID-19 fungal complication, especially in immunocompromised patients. On these lines, we have proposed a temporal convolutional network-based binary classification approach to discover new antifungal molecules in the proteome of plants and animals to accelerate the development of antifungal medications. Although these biomolecules, known as antifungal peptides (AFPs), are part of an organism's intrinsic host defense mechanism, their identification and discovery by traditional biochemical procedures is arduous. Also, the absence of a large dataset on AFPs is also a considerable impediment in building a robust automated classifier. To this end, we have employed the transfer learning technique to pre-train our model on antibacterial peptides. Subsequently, we have built a classifier that predicts AFPs with accuracy and precision of 94%. Our classifier outperforms several state-of-the-art models by a considerable margin. The results of its performance were proven as statistically significant using the Kruskal-Wallis H test, followed by a post hoc analysis performed using the Tukey honestly significant difference (HSD) test. Furthermore, we identified potent AFPs in representative animal (Histatin) and plant (Snakin) proteins using our model. We also built and deployed a web app that is freely available at https://tcn-afppred.anvil.app/ for the identification of AFPs in protein sequences.

Application of machine intelligence for osteoarthritis classification: a classical implementation and a quantum perspective

Osteoarthritis is the most common form of arthritis in the knee that comes with a variation in symptoms’ intensity, frequency and pattern. Knee OA (KOA) is often diagnosed using invasive and expensive methods that can measure changes in joint morphology and function. Early and accurate identification of significant risk factors in clinical data is of vital importance in diagnosing KOA. A machine intelligence approach is proposed here to enable automated, non-invasive identification of risk factors from self-reported clinical data about joint symptoms, disability, function and general health. The proposed methodology was applied to recognize participants with symptomatic KOA or being at high risk of developing KOA in at least one knee. Different machine learning and deep learning algorithms were tested and compared in terms of multiple criteria e.g. accuracy, per class accuracy and execution time. Deep learning was proved to be the most effective in terms of accuracy with classification accuracies up to 86.95%, evaluated on data from the osteoarthritis initiative study. Insights about ten different feature subsets and their effect on classification accuracy are provided. The proposed methodology was also demonstrated in subgroups defined by gender and age. The results suggest that machine intelligence and especially deep learning may facilitate clinical evaluation, monitoring and even prediction of knee osteoarthritis. Apart from the classical implementation of the proposed methodology, a quantum perspective is also discussed highlighting the future application of quantum computers in OA diagnosis.

Deep learning-based cardiovascular image diagnosis: A promising challenge

Artificial intelligence (AI) is becoming a vital concept in medicine leading to a rapid emergence of important tools for medical diagnostics. Now, as a crucial machine learning tool in the field of computer vision, deep learning (DL) is being widely used in medical imaging. Furthermore, as reported in the medical literature, DL has been widely used in medical related research. However, the practical application of DL in clinical diagnosis is relatively small, and it is a new field that may have some challenges. How to effectively perform medical image analysis is a major problem in the field of disease diagnosis, and further diagnostic methods need to be developed. At this stage, DL could be viewed as a black box requiring knowledge of its internal workings, and hence presents some crucial technical challenges that need further methodological development. Thereafter with proper diagnostics, pre-operative computerized simulation planning can be carried out for use of appropriate surgical intervention technology. This paper presents important questions on cardiovascular disease (CVD) diagnostics, using this powerful and yet not adequately understood technology. It discusses issues brought by the paradigm shift of AI vis-à-vis DL in CVD diagnostics, provides possible solutions to potential issues, and envisions the future of the related machine intelligence applications. The discussed problems are dissected into the modular aspects of DL in relation to CVD image classification, segmentation, and detection. A proper perspective on management of these issues is the key to a successful technological implementation of DL in modern medical science.

Separable neural representations of sound sources: Speaker identity and musical timbre

Human listeners can quickly and easily recognize different sound sources (objects and events) in their environment. Understanding how this impressive ability is accomplished can improve signal processing and machine intelligence applications along with assistive listening technologies. However, it is not clear how the brain represents the many sounds that humans can recognize (such as speech and music) at the level of individual sources, categories and acoustic features. To examine the cortical organization of these representations, we used patterns of fMRI responses to decode 1) four individual speakers and instruments from one another (separately, within each category), 2) the superordinate category labels associated with each stimulus (speech or instrument), and 3) a set of simple synthesized sounds that could be differentiated entirely on their acoustic features. Data were collected using an interleaved silent steady state sequence to increase the temporal signal-to-noise ratio, and mitigate issues with auditory stimulus presentation in fMRI. Largely separable clusters of voxels in the temporal lobes supported the decoding of individual speakers and instruments from other stimuli in the same category. Decoding the superordinate category of each sound was more accurate and involved a larger portion of the temporal lobes. However, these clusters all overlapped with areas that could decode simple, acoustically separable stimuli. Thus, individual sound sources from different sound categories are represented in separate regions of the temporal lobes that are situated within regions implicated in more general acoustic processes. These results bridge an important gap in our understanding of cortical representations of sounds and their acoustics.

Examining the misalignment of a linear guideway pair on a feed drive system under different ball screw preload levels with a cost-effective MEMS vibration sensing system

This paper presents an innovative and cost-effective scheme for examining on line the degree of misalignment of a pair of linear guideway rails in a feed drive system. Prototype MEMS-based vibration detection modules were constructed and installed on the ball screw nut and linear guideway block. The ball screw preload was adjusted with four settings by changing ball sizes in this feed drive system. For each preload setting, a linear guideway rail was adjusted using feeler gauges to produce 11 misalignment statuses. The y-axis deviation caused by misalignment along the feed direction (x-axis) was measured with a dial gauge. For each preload and misalignment setting, the vibration signal was ascertained from the installed detection modules while the work table was driven by the controller. By analyzing the vibration signals from the detection module on the linear guideway block, we found that the characteristic frequency in a specific spectrum band could be used as a misalignment index. The characteristic frequencies exhibited a descending trend when the deviation was less than 40μm and an increasing trend when the deviation was 40–120μm. These findings demonstrate the feasibility of using cost-effective sensing units to examine the parallelism or distinguish the deviation between the linear guideway rails without disassembling the work table. This could be valuable for machine intelligence applications in the automation industry.

Salient motion detection using proximal robust principal component analysis method

The recently proposed robust principal component analysis (RPCA) theory and its derived methods have attracted much attention in many computer vision and machine intelligence applications. From a wide view of these methods, independent motion objects are modeled as pixel-wised sparse or structurally sparse outliers from a highly correlated background signal, and all these methods are implemented under an l1 -penalized optimization. Real data experiments reveal that even if l1-penalty is convex, the optimization sometimes cannot be satisfactorily solved, especially when the signal-to-noise ratio is relatively high. In addition, the unexpected background motion (e.g., periodic or stochastic motion) may also be included. We propose a moving object detection method based on a proximal RPCA along with saliency detection. Convex penalties including low-rank and sparse regularizations are substituted with proximal norms to achieve robust regression. After the foreground candidates have been extracted, a motion saliency map using spatiotemporal filtering is constructed. The foreground objects are then filtered out by dynamically adjusting the penalty parameter according to the corresponding saliency values. Evaluations on challenging video clips and qualitative and quantitative comparisons with several state-of-the-art methods demonstrate that the proposed approach works efficiently and robustly.

Angle Modulated Artificial Bee Colony Algorithms for Feature Selection

Optimal feature subset selection is an important and a difficult task for pattern classification, data mining, and machine intelligence applications. The objective of the feature subset selection is to eliminate the irrelevant and noisy feature in order to select optimum feature subsets and increase accuracy. The large number of features in a dataset increases the computational complexity thus leading to performance degradation. In this paper, to overcome this problem, angle modulation technique is used to reduce feature subset selection problem to four-dimensional continuous optimization problem instead of presenting the problem as a high-dimensional bit vector. To present the effectiveness of the problem presentation with angle modulation and to determine the efficiency of the proposed method, six variants of Artificial Bee Colony (ABC) algorithms employ angle modulation for feature selection. Experimental results on six high-dimensional datasets show that Angle Modulated ABC algorithms improved the classification accuracy with fewer feature subsets.

Genetic algorithms for autonomous robot navigation

Engineers and scientists use instrumentation and measurement equipment to obtain information for specific environments, such as temperature and pressure. This task can be performed manually using portable gauges. However, there are many instances in which this approach may be impractical; when gathering data from remote sites or from potentially hostile environments. In these applications, autonomous navigation methods allow a mobile robot to explore an environment independent of human presence or intervention. The mobile robot contains the measurement device and records the data then either transmits it or brings it back to the operator. Sensors are required for the robot to detect obstacles in the navigation environment, and machine intelligence is required for the robot to plan a path around these obstacles. The use of genetic algorithms is an example of machine intelligence applications to modern robot navigation. Genetic algorithms are heuristic optimization methods, which have mechanisms analogous to biological evolution. This article provides initial insight of autonomous navigation for mobile robots, a description of the sensors used to detect obstacles and a description of the genetic algorithms used for path planning.

Intelligent measurement — A view of the state of the art and current trends

The paper examines briefly the state of the art of intelligent measurement on the basis of the proceedings of the IMEKO 1986 Symposium on the subject. It identifies advances in computer-assisted measurement, data conditioning, measurement process control, measurement result output, and measurement data processing as well as application of machine intelligence to measurement interpretation and use in decision support. It identifies the latter as a significant area of future progress. Applications of intelligent measurement in laboratory automation, electrical measurement, mechanical manufacture and associated inspection and clinical medicine are highlighted.