AI, Volume 3, Issue 3 (September 2022) – 13 articles

Societies of the East are often contrasted with those of the West in their stances toward technology. This paper explores these perceived differences in the context of international ethics guidance for artificial intelligence (AI) and robotics. Japan serves as an example of the East, while Europe and North America serve as examples of the West. The paper’s principal aim is to demonstrate that Western values predominate in international ethics guidance and that Japanese values serve as a much-needed corrective. We recommend a hybrid approach that is more inclusive and truly ‘international’. Following an introduction, the paper examines distinct stances toward robots that emerged in the West and Japan, respectively, during the aftermath of the Second World War, reflecting history and popular culture, socio-economic conditions, and religious worldviews. It shows how international ethics guidelines reflect these disparate stances, drawing on a 2019 scoping review that examined 84 international AI ethics documents. These documents are heavily skewed toward precautionary values associated with the West and cite the optimistic values associated with Japan less frequently. Drawing insights from Japan’s so-called ‘moonshot goals’, the paper fleshes out Japanese values in greater detail and shows how to incorporate them more effectively in international ethics guidelines for AI and robotics. Full article

In the current work, a novel method is presented for generating rules for data classification as well as for regression problems. The proposed method generates simple rules in a high-level programming language with the help of grammatical evolution. The method does not depend on any prior knowledge of the dataset; the memory it requires for its execution is constant regardless of the objective problem, and it can be used to detect any hidden dependencies between the features of the input problem as well. The proposed method was tested on a extensive range of problems from the relevant literature, and comparative results against other machine learning techniques are presented in this manuscript. Full article

Machine learning for medical imaging not only requires sufficient amounts of data for training and testing but also that the data be independent. It is common to see highly interdependent data whenever there are inherent correlations between observations. This is especially to be expected for sequential imaging data taken from time series. In this study, we evaluate the use of statistical measures to test the independence of sequential ultrasound image data taken from the same case. A total of 1180 B-mode liver ultrasound images with 5903 regions of interests were analyzed. The ultrasound images were taken from two liver disease groups, fibrosis and steatosis, as well as normal cases. Computer-extracted texture features were then used to train a machine learning (ML) model for computer-aided diagnosis. The experiment resulted in high two-category diagnosis using logistic regression, with AUC of 0.928 and high performance of multicategory classification, using random forest ML, with AUC of 0.917. To evaluate the image region independence for machine learning, Jenson–Shannon (JS) divergence was used. JS distributions showed that images of normal liver were independent from each other, while the images from the two disease pathologies were not independent. To guarantee the generalizability of machine learning models, and to prevent data leakage, multiple frames of image data acquired of the same object should be tested for independence before machine learning. Such tests can be applied to real-world medical image problems to determine if images from the same subject can be used for training. Full article

To obtain more consistent measurements through the course of a wheat growing season, we conceived and designed an autonomous robotic platform that performs collision avoidance while navigating in crop rows using spatial artificial intelligence (AI). The main constraint the agronomists have is to not run over the wheat while driving. Accordingly, we have trained a spatial deep learning model that helps navigate the robot autonomously in the field while avoiding collisions with the wheat. To train this model, we used publicly available databases of prelabeled images of wheat, along with the images of wheat that we have collected in the field. We used the MobileNet single shot detector (SSD) as our deep learning model to detect wheat in the field. To increase the frame rate for real-time robot response to field environments, we trained MobileNet SSD on the wheat images and used a new stereo camera, the Luxonis Depth AI Camera. Together, the newly trained model and camera could achieve a frame rate of 18–23 frames per second (fps)—fast enough for the robot to process its surroundings once every 2–3 inches of driving. Once we knew the robot accurately detects its surroundings, we addressed the autonomous navigation of the robot. The new stereo camera allows the robot to determine its distance from the trained objects. In this work, we also developed a navigation and collision avoidance algorithm that utilizes this distance information to help the robot see its surroundings and maneuver in the field, thereby precisely avoiding collisions with the wheat crop. Extensive experiments were conducted to evaluate the performance of our proposed method. We also compared the quantitative results obtained by our proposed MobileNet SSD model with those of other state-of-the-art object detection models, such as the YOLO V5 and Faster region-based convolutional neural network (R-CNN) models. The detailed comparative analysis reveals the effectiveness of our method in terms of both model precision and inference speed. Full article

Accurately predicting ambient dust plays a crucial role in air quality management and hazard mitigation. Dust emission is a complex, non-linear response to several climatic variables. This study explores the accuracy of Artificial Intelligence (AI) models: an adaptive-network-based fuzzy inference system (ANFIS) and a multi-layered perceptron artificial neural network (mlp-NN), over the Southwestern United States (SWUS), based on the observed dust data from IMPROVE stations. The ambient fine dust (PM2.5) and coarse dust (PM10) concentrations on monthly and seasonal timescales from 1990–2020 are modeled using average daily maximum wind speed (W), average precipitation (P), and average air temperature (T) available from the North American Regional Reanalysis (NARR) dataset. The model’s performance is measured using correlation (r), root mean square error (RMSE), and percentage bias (% BIAS). The ANFIS model generally performs better than the mlp-NN model in predicting regional dustiness over the SWUS region, with r = 0.77 and 0.83 for monthly and seasonal fine dust, respectively. AI models perform better in predicting regional dustiness on a seasonal timescale than the monthly timescale for both fine dust and coarse dust. AI models better predict fine dust than coarse dust on both monthly and seasonal timescales. Compared to precipitation, air temperature is the more important predictor of regional dustiness on both monthly and seasonal timescales. The relative importance of air temperature is higher on the monthly timescale than the seasonal timescale for PM2.5 and vice versa for PM10. The findings of this study demonstrate that the AI models can predict monthly and seasonal fine and coarse dust, based on the limited climatic data, with good accuracy and with potential implications for research in data sparse regions. Full article

Manuscript writing support services using AI technology have become increasingly available in recent years. In keeping with this trend, we need to sort out issues related to authorship in academic writing. Authorship is attached to the contribution of researchers who report innovative research, the originality of which forms the core of their identity. The most important originality is demonstrated in the discussion of study findings. In the discussion section of this paper, we argue that if a researcher uses AI-based manuscript writing support to draft the discussion section, this does not necessarily diminish the researcher’s originality. Rather, AI support may allow the researcher to perform creative work in a more refined fashion. Presumably, selecting which AI support to use or evaluating and properly adjusting AI would still remain an important aspect of research for researchers. It is thus reasonable to view a researcher as a cooperative existence realized through a network of cooperative work that includes the use of AI. Discussions on this topic will be scientifically and socially important as AI technology advances in the future. Full article

Artificial Intelligence (AI) changed our world in various ways. People start to interact with a variety of intelligent systems frequently. As the interaction between human and AI systems increases day by day, the factors influencing their communication have become more and more important, especially in the field of human-agent negotiation. In this study, our aim is to investigate the effect of knowing your negotiation partner (i.e., opponent) with limited knowledge, particularly the effect of familiarity with the opponent during human-agent negotiation so that we can design more effective negotiation systems. As far as we are aware, this is the first study investigating this research question in human-agent negotiation settings. Accordingly, we present a human-agent negotiation framework and conduct a user experiment in which participants negotiate with an avatar whose appearance and voice are a replica of a celebrity of their choice and with an avatar whose appearance and voice are not familiar. The results of the within-subject design experiment show that human participants tend to be more collaborative when their opponent is a celebrity avatar towards whom they have a positive feeling rather than a non-celebrity avatar. Full article

Lung cancer (LC) is the most common cause of cancer-related deaths in the UK due to delayed diagnosis. The existing literature establishes a variety of factors which contribute to this, including the misjudgement of anatomical structure by doctors and radiologists. This study set out to develop a solution which utilises multiple modalities in order to detect the presence of LC. A review of the existing literature established failings within methods to exploit rich intermediate feature representations, such that it can capture complex multimodal associations between heterogenous data sources. The methodological approach involved the development of a novel machine learning (ML) model to facilitate quantitative analysis. The proposed solution, named EMM-LC Fusion, extracts intermediate features from a pre-trained modified AlignedXception model and concatenates these with linearly inflated features of Clinical Data Elements (CDE). The implementation was evaluated and compared against existing literature using F1 score, average precision (AP), and area under curve (AUC) as metrics. The findings presented in this study show a statistically significant improvement (p < 0.05) upon the previous fusion method, with an increase in F-Score from 0.402 to 0.508. The significance of this establishes that the extraction of intermediate features produces a fertile environment for the detection of intermodal relationships for the task of LC classification. This research also provides an architecture to facilitate the future implementation of alternative biomarkers for lung cancer, one of the acknowledged limitations of this study. Full article

In this paper, a hierarchical reinforcement learning (HRL) architecture, namely a “Hierarchical Deep Deterministic Policy Gradient (HDDPG)” has been proposed and studied. A HDDPG utilizes manager and worker formation similar to other HRL structures. However, unlike others, the HDDPG enables sharing an identical environment and state among workers and managers, while a unique reward system is required for each Deep Deterministic Policy Gradient (DDPG) agent. Therefore, the HDDPG allows easy structural expansion with probabilistic action selection of a worker by the manager. Due to its innate structural advantage, the HDDPG has a merit in building a general AI to deal with a complex time-horizon tasks with various conflicting sub-goals. The experimental results demonstrated its usefulness with a manipulator motion planning problem in a dynamic environment, where path planning and collision avoidance conflict each other. The proposed HDDPG is compared with an HAM and a single DDPG for performance evaluation. The result shows that the HDDPG demonstrated more than 40% of reward gain and more than two times the reward improvement rate. Another important feature of the proposed HDDPG is the biased manager training capability. By adding a preference factor to each worker, the manager can be trained to prefer a certain worker to achieve better success rate for a specific objective if needed. Full article

Today’s air traffic management (ATM) system evolves around the air traffic controllers and pilots. This human-centered design made air traffic remarkably safe in the past. However, with the increase in flights and the variety of aircraft using European airspace, it is reaching its limits. It poses significant problems such as congestion, deterioration of flight safety, greater costs, more delays, and higher emissions. Transforming the ATM into the “next generation” requires complex human-integrated systems that provide better abstraction of airspace and create situational awareness, as described in the literature for this problem. This paper makes the following contributions: (a) It outlines the complexity of the problem. (b) It introduces a digital assistance system to detect conflicts in air traffic by systematically analyzing aircraft surveillance data to provide air traffic controllers with better situational awareness. For this purpose, long short-term memory (LSTMs) networks, which are a popular version of recurrent neural networks (RNNs) are used to determine whether their temporal dynamic behavior is capable of reliably monitoring air traffic and classifying error patterns. (c) Large-scale, realistic air traffic models with several thousand flights containing air traffic conflicts are used to create a parameterized airspace abstraction to train several variations of LSTM networks. The applied networks are based on a 20–10–1 architecture while using leaky ReLU and sigmoid activation function. For the learning process, the binary cross-entropy loss function and the adaptive moment estimation (ADAM) optimizer are applied with different learning rates and batch sizes over ten epochs. (d) Numerical results and achievements by using LSTM networks to predict various weather events, cyberattacks, emergency situations and human factors are presented. Full article

Technological problems related to everyday work elements are real, and IT professionals can solve them. However, when they encounter a problem, they must go to a platform where they can detail the category and textual description of the incident so that the support agent understands. However, not all employees are rigorous and accurate in describing an incident, and there is often a category that is totally out of line with the textual description of the ticket, making the deduction of the solution by the professional more time-consuming. In this project, a solution is proposed that aims to assign a category to new incident tickets through their classification, using Text Mining, PLN and ML techniques, to try to reduce human intervention in the classification of tickets as much as possible, reducing the time spent in their perception and resolution. The results were entirely satisfactory and allowed to us determine which are the best textual processing procedures to be carried out, subsequently achieving, in most of the classification models, an accuracy higher than 90%, making its implementation legitimate. Full article

The paper below takes up the question of whether it is possible to transfer the notion of ‘semantic knowledge’—as a human process of making language generate and confer meanings—to machines, which have as one of their properties the capability of handling high amounts of information. This issue is presented in an extended introduction to the paper’s account of and solutions to this intricate problem. Thereafter, the theoretical notion of ‘knowledge’ is considered in its philosophical, and thereby scientific, context, and the basis of its modern import is pointed to being Immanuel Kant’s deliberations on a priori vs. a posteriori knowledge. The author’s solution to the predicament of modern ideas about knowledge is the proposed theory of Occurrence Logic, invented by the author, which abandons truth-values from valid reasoning, and this approach is briefly accounted for. It presupposes a theoretical model of human cognitive systems, and the author has such a model under development which, in the future, may be able to solve the question of what ‘semantic knowledge’ actually is. So far, the theoretical account in this paper points to the critical issue of whether natural language semantics can be grasped as words explaining words or must include the connection between words and objects in the world. The author is in favour of the last option. This leads to the question of the functions of the human brain as the organ connecting words with the outer world. The idea of the so-called ‘predictive brain’ is referred to as a possible solution to the brain/cognition issue, and the paper concludes with a suggestion that an emulation of the interaction between the mentioned cognitive systems may cast some new light on the field of Artificial Intelligence. Full article

Since tiny optical glass is the key component in various optical instruments, more and more researchers have paid attention to automatic defect detection on tiny optical glass in recent years. It remains a challenging problem, as the defects are extremely small. In this paper, we propose a video-based two-stage defect detection network to improve detection accuracy for small defects. Specifically, the detection process is carried out in a coarse-to-fine manner to improve the detection precision. First, the optical glass area is located on the down-sampled version of the input image, and then defects are detected only within the optical glass area with a higher resolution version, which can significantly reduce the false alarming rate. Since the defects may exist on any place of the optical glass, we fuse the results of multiple video frames captured from various perspectives to promote recall rates of the defects. Additionally, we propose an image quality evaluation module based on a clustering algorithm to select video frames with high quality for improving both detection recall and precision. We contribute a new dataset called OGD-DET for tiny-scale optical glass surface defect detection experiments. The datasets consist of 3415 images from 40 videos, and the size of the defect area ranges from 0.1 mm to 0.53 mm, 2 to 7 pixels on images with a resolution of 1536 × 1024 pixels. Extensive experiments show that the proposed method outperforms the state-of-the-art methods in terms of both accuracy and computation cost. Full article