World Scientific Annual Review of Artificial Intelligence

With advanced imaging, sequencing, and profiling technologies, multiple omics data become increasingly available and hold promises for many healthcare applications such as cancer diagnosis and treatment. Multimodal learning for integrative multi-omics analysis can help researchers and practitioners gain deep insights into human diseases and improve clinical decisions. However, several challenges are hindering the development in this area, including the availability of easily accessible open-source tools. This survey aims to provide an up-to-date overview of the data challenges, fusion approaches, datasets, and software tools from several new perspectives. We identify and investigate various omics data challenges that can help us understand the field better. We categorize fusion approaches comprehensively to cover existing methods in this area. We collect existing open-source tools to facilitate their broader utilization and development. We explore a broad range of omics data modalities and a list of accessible datasets. Finally, we summarize future directions that can potentially address existing gaps and answer the pressing need to advance multimodal learning for multi-omics data analysis.

With the development of information technology, users usually suffer from the problem of information overload. To solve this problem, recommendation systems (RSs) are widely explored to help the user find potential interest in the overloaded information. Since the recommendation scenarios become more and more abundant, distinct recommendation tasks are accordingly proposed by the researchers. In this survey, we comprehensively investigate various recommendation domains and scenarios, and provide a classification scheme for RSs, aiming to have considerable insight for further research. This survey is supposed to give a systematic understanding of the key components in RSs and summarize the diversified characteristics in each domain or scenario.

Recently, many machine learning models have been proposed to understand and analyze Programming Languages (PLs). While there are some similarities between PLs and Natural Language Processing (NLP), the former one has its own unique challenges. In this survey, we investigate current approaches tackling representation learning of codes and associated downstream tasks that can be solved with them. We present and compare the state-of-the-art models specifically designed for embedding PLs in low-dimensional space, and demonstrate how these embedding methods are related to representation learning approaches in NLP. We also compare benchmark experiments on multiple code-related tasks and evaluate the models for each specific application.

Cross-lingual text classification is a challenging task in natural language processing. The objective is to build accurate text classification models for low-resource languages by transferring the knowledge learned from high-resource languages. The task has been studied since 2003 and has attracted significantly growing attention in the last decade due to the success of deep learning models in natural language processing. Many new methods have been proposed to address the challenges in cross-lingual text classification. Meanwhile, cross-lingual fake news detection is one of the most important applications of cross-lingual text classification. It has already created significant social impacts on alleviating the infodemic problem in low-resource languages. The research works on cross-lingual text classification and cross-lingual fake news detection have been growing rapidly in recent years. Therefore, a comprehensive survey is imperative to summarize existing algorithms for cross-lingual text classification and explain the connections among them. This paper systematically reviews research works on cross-lingual text classifications and their applications in cross-lingual fake news detection. We categorize the evolution of cross-lingual text classification methods into four phases: (1) Traditional text classification models with translation; (2) Cross-lingual word embedding-based methods, (3) Pretraining then finetuning-based methods, and (4) Pretraining then prompting-based methods. We first discuss and analyze the representative methods in each phase in detail. Second, we provide a detailed review of their applications in the emerging fake news detection problem. Finally, we explore the potential issues of this open problem and also discuss possible future directions.

Evidence-based Medicine (EBM) is facing a new challenge in applying Artificial Intelligence (AI) for better services. Here the fundamental question is: can AI be used as a human intelligence amplifier to make better evidence-based medical decisions? This argument can only be answered if AI-supported decisions are compared and contrasted with those of human experts, and if the implementation of AI-enabled Evidence-Based Medicine (AiEBM) can be explained or justified using the available evidence and experience. In this paper, we propose AiEBM as a human-centered approach to be integrated into processes of healthcare systems, thereby increasing the credibility of the clinical decision support systems. We provide insights into the applicability of AI and its challenges and opportunities in EBM. The AaaS (AI as a Service) and high-order reasoning facilitated by machine learning and AI for medical decisions will be discussed.

The impact of climate change on the environment has become increasingly visible today, and foreseeing future climate events, which involves long-term prediction of climate variables (e.g., temperature, wind speed, precipitation, etc.) at a local small scale in a local region, is crucial for disaster risk management. General Circulation Models (GCMs) allow for the simulation of multiple climate variables, decades into the future (often till the year 2100). GCM simulations, however, are at a global large scale (from 100 km to 600 km) and are too coarse to monitor climate change at the local small scale. Statistical downscaling approaches are often applied to the GCM simulations to allow the evaluation of the GCM outputs at the local scale. Machine learning-based techniques are popular approaches for statistical downscaling. In this paper, we provide an overview of GCM downscaling with machine learning and present a case study that leverages deep learning to downscale weekly averages of the daily minimum and maximum temperatures in the Hackensack–Passaic watershed in New Jersey.

With a growing interest in autonomous vehicles’ operation, there is an equally increasing need for efficient anticipatory gesture recognition systems for human-vehicle interaction. Existing gesture recognition algorithms have been primarily restricted to historical data. In this paper, we propose a novel Context-and-Gap-aware Pose Prediction Framework (CGAP2), which predicts the future pose data for anticipatory recognition of gestures in an online fashion. CGAP2 implements an encoder–decoder architecture paired with a pose prediction module to anticipate future frames followed by a shallow classifier. CGAP2’s pose prediction module uses 3D convolutional layers and is dependent on the number of pose frames supplied, the time difference between subsequent pose frames, and the number of predicted pose frames. The performance of CGAP2 is evaluated on the Human3.6M dataset with the MPJPE metric. For the pose prediction of 15 frames in advance, an error of 79 mm is achieved. The pose prediction module consists of only 26M parameters and can run at 50 FPS on the NVIDIA TITAN RTX. CGAP2 has a 1-s time advantage compared to other gesture recognition systems, which can be crucial for autonomous vehicles. Furthermore, this work formalizes the problem of anticipatory gesture recognition by introducing two hyper-parameters namely context and gap and a relationship is established between the two using carefully designed experiments in ablation testing. As per our knowledge, this work is the first to tackle all the challenges faced by anticipatory gesture recognition and also the first to post a performance benchmark on Human3.6M dataset.

Facial expressions (FEs) are one of the most preeminent means of conveying one’s emotions and are pivotal to nonverbal communication. Potential applications in a wide range of areas in computer vision have lent a strong impetus to research in the domain of automatic facial expression recognition. This work discusses the effectiveness of two optical flow-based features for modeling the FEs associated with prototypic emotions based on the pattern of nonrigid deformable motion of facial components occurring during their portrayal. The discernible motion patterns are categorized into distinct discrete classes with the descriptive features indicating the global spatial distribution of deformation derived from the dense optical flow field associated with emotional and neutral face images. Results obtained with evaluation on images and video clips taken from Extended Cohn-Kanade, Japanese Female Facial Expressions, and Dynamic Karolinska Directed Emotional Faces datasets with multi-class support vector machine and $k$-nearest neighbor classifiers are competent with the state-of-the-art techniques and concordant with empirical psychological studies in emotion science.

Adversarial examples (AEs) are malicious test-data samples (typically images) generated by applying carefully calculated perturbations to clean samples. The added perturbations are usually human-imperceptible but the AEs can fool a machine learning (ML) model to make misclassifications. Although multiple methods were proposed to generate AEs, the ability to generalize is very limited; that is, they easily overfit to their source, single, white-box ML models and the generated AEs rarely work for other models. In this paper, we propose a black-box attack approach that crafts transferable AEs that can attack a wide range of ML models without knowing those model details. Our novel method consists of an elastic momentum (EM) that expedites gradient descent to avoid early overfitting, and a random erasure (RE) technique that increases the diversity of perturbations and reduces gradient fluctuations. Our method can be applied to any gradient-based attacks to make those attacks become more transferable. We evaluate our proposed method by attacking seven state-of-the-art (SOTA) deep learning models and comparing against five SOTA attacks; we also attack nine advanced defense mechanisms that are integrated into the above models. Our results demonstrate significant improvement on the attack success rate (ASR) and transferability when using our method alone, and that it can also be easily applied to other baseline methods (which are gradient-based) to substantially improve their performance.

Graph Neural Networks (GNNs) have become the de facto standard for representation learning on topological graphs, which usually derive effective node representations via message passing from neighborhoods. Although GNNs have achieved great success, previous models are mostly confined to static and homogeneous graphs. However, there are multiple dynamic interactions between different-typed nodes in real-world scenarios like academic networks and e-commerce platforms, forming temporal heterogeneous graphs (THGs). Limited work has been done for representation learning on THGs and the challenges are in two aspects. First, there are abundant dynamic semantics between nodes while traditional techniques like meta-paths can only capture static relevance. Second, different semantics on THGs have often mutually evolved with each other over time, making it more difficult than dynamic homogeneous graph modeling. To address this problem, here we propose the Dynamic Meta-path guided temporal heterogeneous Graph Neural Networks (DyMGNNs). To handle the dynamic semantics, we introduce the concept of dynamic meta-path which is a common base for temporal semantic search engines, and then adopt the temporal importance sampling to extract neighborhoods with temporal bias. Focusing on mutual evolution, we design the heterogeneous mutual evolution attention mechanism, which can model the fine-grained interplay of semanticlevel preferences for each node. Extensive experiments on three real-world datasets for node classification and temporal link prediction demonstrate that our method consistently outperforms state-of-the-art alternatives.

Frame identification is a crucial and challenging task in frame semantic parsing, where the objective is to determine the most appropriate frame for a given target within its context. Existing approaches typically classify each target individually, overlooking the potential interactions among different targets. To address these limitations, we propose a novel two-step prediction framework that encompasses both local and global perspectives for target identification. Additionally, we leverage the frame ontology graph to enhance the interactions among multiple targets by incorporating rich frame ontology knowledge. Moreover, we introduce a dynamic masking strategy during model training. This strategy encourages our model to adopt a global view during prediction, avoiding local optimization. Experimental results demonstrate the superiority of our model compared to previous approaches on FrameNet1.5, and it achieves competitive performance on FrameNet1.7. Furthermore, supplementary experiments and analyses provide additional evidence of the effectiveness of our proposed model.