Search

Multimodal emotion recognition, that is, emotion recognition uses machine learning to generate multi-modal features on the basis of videos which has become a research hotspot in the field of artificial intelligence. Traditional multi-modal emotion recognition method only simply connects multiple modalities, and the interactive utilization rate of modal information is low, and it cannot reflect the real emotion under the conflict of modal features well. This article first proves that effective weighting can improve the discrimination between modalities. Therefore, this paper takes into account the importance differences between multiple modalities, and assigns weights to them through the importance attention network. At the same time, considering that there is a certain complementary relationship between the modalities, this paper constructs an attention network with complementary modalities. Finally, the reconstructed features are fused to obtain a multi-modal feature with good interaction. The method proposed in this paper is compared with traditional methods in public datasets. The test results show that our method is accurate in It performs well in both the rate and confusion matrix metrics. [ABSTRACT FROM AUTHOR]

• Many ML algorithms are used to identify the main determinants of a bank default. • We use of a graph neural network that has never been used in a financial context. • We obtain a balanced dataset by customizing the heuristic oversampling method. • Like previous literature, we show that neural network outperforms other approaches. • We include, for the first time, competition among the possible default determinants. The financial sector is very interested in Artificial Intelligence due to the opportunities that it offers, especially those related to methods of machine-learning. The aim of this paper is to employ a variety of machine-learning algorithms to identify the main determinants of bank default and to understand the impact of each variable on it. Bank default is one of the most studied topics in financial literature because of the severity of its consequences on the whole economic system. However, little attention has been paid to the identification of the major determinants of bank failures via machine-learning approaches. This paper employs several machine-learning algorithms, including a graph neural network that has never been used in a financial context. Another novelty is the implementation of a balanced dataset by customising the heuristic oversampling method based on k-means and synthetic minority over-sampling technique. This paper also deals with the inclusion of competition among the possible default determinants. The dataset consists of all the banks in the Euro Area in the period 2018–2020. The results obtained are useful from both micro- and macro-economic points of view. [ABSTRACT FROM AUTHOR]

This paper introduces the Ulimisana Optimisation Algorithm enabled Population Based Training (PBT-UOA) framework which allows for hyperparameters to be fine-tuned using a population based meta-heuristic algorithm at the same time as parameters are being optimised. Models are trained until near-convergence on the updated hyperparameters and the parameters of the best performing model are shared to warm start the other models in the next hyperparameter tuning iteration. In the PBT-UOA, all models are trained using the same dataset. This framework performed better than the Bayesian Optimisation algorithm. This paper also introduces the Ulimisana Optimisation Algorithm enabled Federated Learning (FL-UOA) framework which is an extension of the PBT-UOA. This framework is introduced to address the challenges of scattered datasets and privacy that is presented by the increase in connected end-devices. The FL-UOA learns on local data in scattered end-devices without sending datasets to a central server. The training datasets in local end-devices are used to evaluate models trained in other end-devices. The performance metrics are used to update the Social Trust Network (STN) of the FL-UOA framework. The FL-UOA outperformed the classic Federated Learning framework. This STN updating technique was tested in Machine Learning (ML) Unfairness to see how well it functioned as a regularisation term. This was achieved by training different models on subsets that contained datasets representing only specific sensitive groups. Results showed that by updating the hyperparameters while learning the parameters on the dataset scattered across different devices, the FL-UOA, takes advantage of diversified learning and reduces the ML Unfairness for models trained on group specific datasets. [ABSTRACT FROM AUTHOR]

How to derive an investor's preference is vital for investment advisors and online lending platforms for targeted marketing strategies, e.g., market segmentation and financial product recommendation. However, investor preference analysis usually depends on judgments from human investment experts, which are inherently subjective and costly. Intelligent investment advisors (or Robo-advisors), supported by cutting-edge technologies such as machine learning and artificial intelligence, are established to relieve these pending issues. This paper employs an online optimization framework to obtain investors' preferences for further financial product recommendations. This proposed method allows us to update the investor's preference for newly-arriving data sets and tackle the situation where plenty of missing values in investors' records are present. Unlike the black-box-like machine learning approach, our method can provide more managerial implications regarding why one financial product/service is preferred. Real-world data set from an online financial platform is used to compare the existing approaches and shows the stronger and more stable performance of our method when facing different data-missing types and situations with different missing degrees, followed by a recommendation system application. [ABSTRACT FROM AUTHOR]

Explainable artificial intelligence (XAI) is gradually becoming a key component of many artificial intelligence systems. However, such pursuit of transparency may bring potential privacy threats to the model confidentially, as the adversary may obtain more critical information about the model. In this paper, we systematically study how model decision explanations impact model extraction attacks, which aim at stealing the functionalities of a black-box model. Based on the threat models we formulated, an XAI-aware model extraction attack (XaMEA), a novel attack framework that exploits spatial knowledge from decision explanations is proposed. XaMEA is designed to be model-agnostic: it achieves considerable extraction fidelity on arbitrary machine learning (ML) models. Moreover, we proved that this attack is inexorable, even if the target model does not proactively provide model explanations. Various empirical results have also verified the effectiveness of XaMEA and disclosed privacy leakages caused by decision explanations. We hope this work would highlight the need for techniques that better trade off the transparency and privacy of ML models. • We propose XaMEA, three XAI-aware model extraction attack architectures. • We further carry out XAI-aware model extraction attacks against non-explanation target models. • We evaluate the attack effectiveness of XaMEA with an exhaustive set of experiments. [ABSTRACT FROM AUTHOR]

This paper presents a method called Interval Incremental Learning (IIL) to capture spatial and temporal patterns in uncertain data streams. The patterns are represented by information granules and a granular rule base with the purpose of developing explainable human-centered computational models of virtual and physical systems. Fundamentally, interval data are either included into wider and more meaningful information granules recursively, or used for structural adaptation of the rule base. An Uncertainty-Weighted Recursive-Least-Squares (UW-RLS) method is proposed to update affine local functions associated with the rules. Online recursive procedures that build interval-based models from scratch and guarantee balanced information granularity are described. The procedures assure stable and understandable rule-based modeling. In general, the model can play the role of a predictor, a controller, or a classifier, with online sample-per-sample structural adaptation and parameter estimation done concurrently. The IIL method is aligned with issues and needs of the Internet of Things, Big Data processing, and eXplainable Artificial Intelligence. An application example concerning real-time land-vehicle localization and tracking in an uncertain environment illustrates the usefulness of the method. We also provide the Driving Through Manhattan interval dataset to foster future investigation. [ABSTRACT FROM AUTHOR]

• Explainable framework for meta-learning. • Efficiency and high causality. • Intervention and counterfactual. With the growing convergence of artificial intelligence and daily life scenarios, the application scenarios for intelligent decision methods are becoming increasingly complex. The development of various machine learning algorithms has benefited all disciplines of study, but choosing which algorithm is most suitable for a certain problem among a large number of algorithms is a challenge that every field must overcome. Another challenge at the practical application level is that machine learning algorithms currently trained with large amounts of data are primarily black-box and uninterpretable. This indicates that these methods pose potential risks and are difficult to rely on, thus hindering their application in sensitive fields such as finance and healthcare. The first challenge can be overcome by using meta-learning to combine data and prior knowledge to efficiently and automatically select the machine learning models. The second challenge remains to be addressed due to the lack of interpretability of traditional meta-learning techniques and deficiencies in transparency and fairness. Achieving the interpretability of meta-learning in autonomous algorithm selection for classification is crucial to balance the need for high accuracy and transparency of machine learning models in practical application scenarios. This paper proposes EFFECT , an interpretable meta-learning framework that can explain the recommendation results of meta-learning algorithm selection and provide a more complete and accurate explanation of the recommendation algorithm's performance on specific datasets combined with business scenarios. Extensive experiments have demonstrated the validity and correctness of this framework. [ABSTRACT FROM AUTHOR]

• Explainable framework for meta-learning. • Efficiency and high causality. • Intervention and counterfactual. With the growing convergence of artificial intelligence and daily life scenarios, the application scenarios for intelligent decision methods are becoming increasingly complex. The development of various machine learning algorithms has benefited all disciplines of study, but choosing which algorithm is most suitable for a certain problem among a large number of algorithms is a challenge that every field must overcome. Another challenge at the practical application level is that machine learning algorithms currently trained with large amounts of data are primarily black-box and uninterpretable. This indicates that these methods pose potential risks and are difficult to rely on, thus hindering their application in sensitive fields such as finance and healthcare. The first challenge can be overcome by using meta-learning to combine data and prior knowledge to efficiently and automatically select the machine learning models. The second challenge remains to be addressed due to the lack of interpretability of traditional meta-learning techniques and deficiencies in transparency and fairness. Achieving the interpretability of meta-learning in autonomous algorithm selection for classification is crucial to balance the need for high accuracy and transparency of machine learning models in practical application scenarios. This paper proposes EFFECT , an interpretable meta-learning framework that can explain the recommendation results of meta-learning algorithm selection and provide a more complete and accurate explanation of the recommendation algorithm's performance on specific datasets combined with business scenarios. Extensive experiments have demonstrated the validity and correctness of this framework. [ABSTRACT FROM AUTHOR]

• Explanation Sets, a new framework that unifies counterfactuals and semifactuals. • Counterfactuals and semifactuals are defined in terms of a similarity measure. • Restrictions and preferences over the explanations are defined using a neighborhood. • A taxonomy for set-based representations. Explainable Machine Learning (ML) is an emerging field of Artificial Intelligence that has gained popularity in the last decade. It focuses on explaining ML models and their predictions, enabling people to understand the rationale behind them. Counterfactuals and semifactuals are two instances of Explainable ML techniques that explain model predictions using other observations. These techniques are based on the comparison between the observation to be explained and another one. In counterfactuals, their prediction is different, and in semifactuals, it is the same. Both techniques have been studied in the Social Sciences and Explainable ML communities, and they have different use cases and properties. In this paper, the Explanation Set framework, an approach that unifies counterfactuals and semifactuals, is introduced. Explanation Sets are example-based explanations defined in a neighborhood where most observations satisfy a grouping measure. The neighborhood allows defining and combining restrictions. The grouping measure determines if the explanations are counterfactuals (dissimilarity) or semifactuals (similarity). Besides providing a unified framework, the major strength of the proposal is to extend these explanations to other tasks such as regression by using an appropriate grouping measure. The proposal is validated in a regression and classification task using several neighborhoods and grouping measures. [ABSTRACT FROM AUTHOR]

In response to fighting COVID-19 pandemic, researchers in machine learning and artificial intelligence have constructed some medical knowledge graphs (KG) based on existing COVID-19 datasets, however, these KGs contain a considerable amount of semantic relations which are incomplete or missing. In this paper, we focus on the task of knowledge graph embedding (KGE), which serves an important solution to infer the missing relations. In the past, there have been a collection of knowledge graph embedding models with different scoring functions to learn entity and relation embeddings published. However, these models share the same problems of rarely taking important features of KG like attribute features, other than relation triples, into account, while dealing with the heterogeneous, complex and incomplete COVID-19 medical data. To address the above issue, we propose a graph feature collection network (GFCNet) for COVID-19 KGE task, which considers both neighbor and attribute features in KGs. The extensive experiments conducted on the COVID-19 drug KG dataset show promising results and prove the effectiveness and efficiency of our proposed model. In addition, we also explain the future directions of deepening the study on COVID-19 KGE task. [ABSTRACT FROM AUTHOR]

The rapid integration of intelligent processes and methods into information systems in the Artificial Intelligence (AI) era has led to a substantial shift towards autonomous software decision-making. This evolution necessitates robust human oversight, especially in critical domains like Healthcare, Education, and Energy. Human trust in AI plays a vital role in influencing decision-making processes of users interacting with AI. This paper presents VIRTSI (V ariability and I mpact of R eciprocal T rust S tates towards I ntelligent systems), a novel rigorous computational model for human-AI Interaction. VIRTSI simulates human trust states, spanning from overtrust to distrust, through user modelling. It comprises: 1. A trust dynamics representational model based on Deterministic Finite State Automata (DFAs), illustrating transitions among cognitive trust states in response to AI-generated replies. 2. A trust evaluation model based on Confusion Matrices, originating from machine learning and Accuracy Metrics, providing a quantitative framework for analysing human trust dynamics. As a result, this is the first time that trust dynamics have been thoroughly traced in a representational model and a method has been developed to assess the impact of possibly harmful states like overtrust and distrust. An empirical study on the recently launched Large Language Model of generative AI, ChatGPT (version 3.5), provides a radical underexplored AI-generated platform for evaluating the human-AI interaction through VIRTSI. The study involved 1200 interactions of real users as well as AI experts together with experts in two very different domains of evaluation, namely software engineering and poetry. This study traces trust dynamics and the emerging human-AI interaction, in concrete examples of real user synergies with generative AI. The research reveals the vital role of maintaining normal trust states for optimal human-AI interaction and that both AI and human users need further steps towards this goal. The real-world implications of this research can guide the creation and evaluation of user interfaces with AI and the incorporation of functionalities in the development of generative AI chatbots in terms of trust by providing a new rigorous DFA representational method of trust dynamics and a corresponding new perspective of confusion matrix evaluation method of the dynamics' impact in the efficiency of human-AI dialogues. [ABSTRACT FROM AUTHOR]

The use of artificial intelligence for decision support and automation has shown tremendous potential in many areas. The ability to explain the decisions made by a machine learning algorithm is fundamental to facilitating the widespread use of this type of tool. There are many important real-world problems where the cost of the decisions depends on the characteristics of each example: these are called example-dependent cost (EDC) problems. For this type of classification problem, an appropriate formulation that takes into account the decision costs is fundamental both for the design of the classifier and for the explanation of its decisions. In this paper, we propose COCOA, an explanation method designed for EDC problems based on a Bayesian discriminant. The proposed method can provide counterfactual samples generated by considering decision costs. The COCOA method provides valid and plausible counterfactuals with a high success rate, which can be actionable, diverse, and sparse, achieving a remarkable improvement in terms of cost over five state-of-the-art methods on six real-world datasets. • COCOA provides actionable, plausible, sparse and diverse counterfactuals. • Decision costs are taken into account when generating counterfactuals. • The obtained counterfactuals improve cost savings. • The method is presented for MLPs, but other architectures can be used. [ABSTRACT FROM AUTHOR]

Feature importance and interaction are among the main issues in explainable artificial intelligence or interpretable machine learning. To measure feature importance and interaction, several methods, such as H-statistic and partial dependency, have been proposed. However, it is difficult to understand the practical implications of importance and interaction. In this paper, a new method for measuring feature importance and interaction is proposed. For the classification model, we observed correctly predicted cases in a predictive model and grouped them according to the characteristics of the cases. We derived a method for feature importance and interaction from group information. For the regression model, we grouped cases according to the change in the size of the prediction error. The proposed method supports the same rationale for feature importance and interaction. It also supports the decomposition of feature importance to feature power and feature interactions. To implement the proposed method, three visualization tools, including a feature interaction graph, are implemented. Through the proposed work, we can better understand the working mechanism of a predictive model. [ABSTRACT FROM AUTHOR]

Progress in the field of Non-Intrusive Load Monitoring (NILM) has been attributed to the rise in the application of artificial intelligence. Nevertheless, the ability of energy disaggregation algorithms to disaggregate different appliance signatures from aggregated smart grid data poses some privacy issues. This paper introduces a new notion of disclosure risk termed energy disaggregation risk. The performance of Sequence-to-Sequence (Seq2Seq) NILM deep learning algorithm along with three activation extraction methods are studied using two publicly available datasets. To understand the extent of disclosure, we study three inference attacks on aggregated data. The results show that Variance Sensitive Thresholding (VST) event detection method outperformed the other two methods in revealing households' lifestyles based on the signature of the appliances. To reduce energy disaggregation risk, we investigate the performance of two privacy-preserving mechanisms based on microaggregation and Discrete Fourier Transform (DFT). Empirically, for the first scenario of inference attack on UK-DALE, VST produces disaggregation risks of 99%, 100%, 89% and 99% for fridge, dish washer, microwave, and kettle respectively. For washing machine, Activation Time Extraction (ATE) method produces a disaggregation risk of 87%. We obtain similar results for other inference attack scenarios and the risk reduces using the two privacy-protection mechanisms. [ABSTRACT FROM AUTHOR]

Technology has grown rapidly in recent years, and new solutions that rely on Machine Learning (ML) and Artificial Intelligence (AI) are introduced every day. With such fast-paced advancement, inspecting and fully comprehending how given models make decisions is becoming problematic. The complex decision-making process of these models has become a black box, making it challenging to unravel how they work; therefore, eXplainable Artificial Intelligence (XAI) methods are crucial for further development. This paper discusses how state-of-the-art techniques determine classifications and why they need to be revised to understand the prediction-generating process fully. It compares those existing solutions with the new method called Principal Image Sections Mapping - PRISM, which relies on Principal Component Analysis and allows visualising the most significant features recognised by a given Convolutional Neural Network. PRISM is implemented in a piece of software called TorchPRISM that can generate and present the clustering based on the method's output. The result can indicate ambiguous classes discrimination; thus, the possibility of automating the output analysis process is also discussed. The paper's main objective is to examine how PRISM enhances the current understanding of the decision-making process and introduce a tool that can facilitate analysing the output. PRISM implementation (TorchPRISM) can be found in the public GitHub repository: https://github.com/szandala/TorchPRISM • PRISM method can explain CNN's decisions in an understandable way for various users. • PRISM uses PCA to point features and highlights them using an RGB-coloured image mask. • PRISM's effectiveness is assessed based on qualitative and quantitative measures. • TorchPRISM software allows the identification of ambiguous classes in large datasets. • PRISM overcomes the deficiencies of existing eXplainable AI and saliency map methods. [ABSTRACT FROM AUTHOR]

• The stochastic separation theorems are applied to develop industrial computer vision systems. • New class of algorithms are presented and verified for removing AI errors. • This class of algorithms is based on stochastic separation theorems combined with clustering. • The algorithms are tested on a relevant real-life problem. In this paper we present theory and algorithms enabling classes of Artificial Intelligence (AI) systems to continuously and incrementally improve with a priori quantifiable guarantees – or more specifically remove classification errors – over time. This is distinct from state-of-the-art machine learning, AI, and software approaches. The theory enables building few-shot AI correction algorithms and provides conditions justifying their successful application. Another feature of this approach is that, in the supervised setting, the computational complexity of training is linear in the number of training samples. At the time of classification, the computational complexity is bounded by few inner product calculations. Moreover, the implementation is shown to be very scalable. This makes it viable for deployment in applications where computational power and memory are limited, such as embedded environments. It enables the possibility for fast on-line optimisation using improved training samples. The approach is based on the concentration of measure effects and stochastic separation theorems and is illustrated with an example on the identification faulty processes in Computer Numerical Control (CNC) milling and with a case study on adaptive removal of false positives in an industrial video surveillance and analytics system. [ABSTRACT FROM AUTHOR]

Accurate estimation of energy savings is crucial for the effective implementation of energy conservation measures (ECMs). Simultaneously, the integration of Artificial Intelligence (AI) has revolutionized software engineering by imbuing software with intelligent capabilities and autonomy. In this paper, we propose an ensemble model for precisely estimating baseline energy consumption within the realm of AI-empowered autonomous software. The ensemble model combines predictions from three tree-based Machine Learning models, namely Random Forest, XGBoost, and LightGBM. Notably, our model emphasizes the provision of explainability, granting transparency and insights into the key factors influencing baseline energy consumption. To validate its effectiveness, we conduct experimental evaluations on a diverse cluster of real-world buildings in Latvia. The results demonstrate the superiority of our proposed ensemble model over individual models and even a deep learning network tailored for energy consumption estimation. These findings underscore the efficacy of AI-empowered models in the energy sector, offering a robust and interpretable solution for estimating energy savings. [ABSTRACT FROM AUTHOR]

Time series are a kind of streaming data, which are chaotic and sequential. As real-world time series data are often not available at once and drift with time growth, Incremental Learning (IL) is well suited for Time Series Prediction (TSP). Most previous incremental TSP algorithms are limited by the assumption of data balance. However, real-world time series data are often unbalanced, with long-tailed distribution and other characteristics resulting in the failure of IL algorithms. In this paper, a balanced-driven Active Learning (AL) strategy is proposed to deal with data imbalance problems in IL processes. What's more, by integrating the advantages of Deep Learning (DL) and the Broad Learning System (BLS), a novel Deep-Broad Learning (DeepBL) network with its incremental learning algorithm is proposed. The proposed A ctive Learning-based I ncremental Deep - B road L earning (AI_DeepBL) algorithm is applied to real-world univariate and multivariate time series datasets and achieves superior performance compared with classical and state-of-the-art TSP algorithms. [ABSTRACT FROM AUTHOR]

In recent years, deep learning has attracted an increasing amount of attention in machine learning and artificial intelligence areas. Currently, many deep learning network-related architectures such as deep neural networks (DNNs), convolutional neural network (CNN), wavelet scattering network (ScatNet) and principal component analysis network (PCANet) have been proposed. The most effective network is PCANet, which has achieved promising performance in image classification, such as for face, object and handwritten digit recognition. PCANet can only handle data that are represented by single-view features. In this paper, we present a canonical correlation analysis network (CCANet) to address image classification, in which images are represented by two-view features. The CCANet learns two-view multistage filter banks by a canonical correlation analysis (CCA) method and constructs a cascaded convolutional deep network. Then, we incorporate filters with binaryzation and block-wise histogram processes to form the final depth structure. In addition, we introduce a variation of CCANet—dubbed RandNet-2—in which the filter banks are randomly generated. Extensive experiments are conducted using the ETH-80, Yale-B, and USPS databases for object classification, face classification and handwritten digits classification, respectively. The experimental results demonstrate that the CCANet algorithm is more effective than PCANet, RandNet-1 and RandNet-2. [ABSTRACT FROM AUTHOR]

Imperfection of information is a part of our daily life; however, it is usually ignored in learning based on evolutionary approaches. In this paper we develop an Imperfect Evolutionary System that provides an uncertain and chaotic imperfect environment that presents new challenges to its habitants. We then propose an intelligent methodology which is capable of learning in such environments. Detecting changes and adapting to the new environment is crucial to exploring the search space and exploiting any new opportunities that may arise. To deal with these uncertain and challenging environments, we propose a novel change detection strategy based on a Particle Swarm Optimization system which is hybridized with an Artificial Neural Network. This approach maintains a balance between exploitation and exploration during the search process. A comparison of approaches using different Particle Swarm Optimization algorithms show that the ability of our learning approach to detect changes and adapt as per the new demands of the environment is high. [ABSTRACT FROM AUTHOR]

In incremental learning techniques, learning occurs continuously over time and does not cease once available data have been exhausted. Such techniques are useful in cases where problem data may be acquired in small quantities over time. This paper presents an incremental neural network called the evolving Probabilistic Neural Network. The main advantage of this technique lies in its adaptive architecture, which adjusts to data distributions. This method requires that each training sample be used only once throughout the training phase without being reprocessed. The technique is flexible and offers a simplified structure while maintaining performance levels comparable to those of other techniques. Experiments were conducted using publicly available benchmark data sets. These experiments show that overall, the proposed model achieves a quality of response that is comparable to those of the best techniques evaluated, and its structure size and classification time were as low as those of less complex techniques. These results indicate that the proposed model achieves a satisfactory balance between efficiency and efficacy. [ABSTRACT FROM AUTHOR]

Modeling dynamical systems is a commonly used technique to solve partially observable problems in the artificial intelligence field. Predictive state representations (PSRs) have been proposed as an alternative to partially observable Markov decision processes (POMDPs) to model dynamical systems. Although POMDPs and PSRs provide general frameworks to solve partially observable problems, they rely heavily on a known and accurate model of the environment. However, in real world applications it is extremely difficult to build an accurate model. In this paper, we propose an algorithm to solve partially observable problems using an inaccurate PSR model which is learned from samples. The proposed algorithm can also improve the accuracy of the learned model. Given the inaccurate PSR model, the PSR state is identified firstly. Then the traditional Markov decision processes (MDP) techniques are used to solve the partially observable problem. Furthermore, the learned model, which may get off-track as often happens when the model is learned from samples, can be reset. The effectiveness of our proposed algorithm is demonstrated based on a standard set of POMDP test problems. [ABSTRACT FROM AUTHOR]