Papers by Andrea Apicella
arXiv (Cornell University), Jun 9, 2023
In the context of classification problems, Deep Learning (DL) approaches represent state of art. ... more In the context of classification problems, Deep Learning (DL) approaches represent state of art. Many DL approaches are based on variations of standard multi-layer feed-forward neural networks. These are also referred to as deep networks. The basic idea is that each hidden neural layer accomplishes a data transformation which is expected to make the data representation "somewhat more linearly separable" than the previous one to obtain a final data representation which is as linearly separable as possible. However, determining the appropriate neural network parameters that can perform these transformations is a critical problem. In this paper, we investigate the impact on deep network classifier performances of a training approach favouring solutions where data representations at the hidden layers have a higher degree of linear separability between the classes with respect to standard methods. To this aim, we propose a neural network architecture which induces an error function involving the outputs of all the network layers. Although similar approaches have already been partially discussed in the past literature, here we propose a new architecture with a novel error function and an extensive experimental analysis. This experimental analysis was made in the context of image classification tasks considering four widely used datasets. The results show that our approach improves the accuracy on the test set in all the considered cases.
Scientific Reports, Apr 7, 2022
A wearable system for the personalized EEG-based detection of engagement in learning 4.0 is propo... more A wearable system for the personalized EEG-based detection of engagement in learning 4.0 is proposed. In particular, the effectiveness of the proposed solution is assessed by means of the classification accuracy in predicting engagement. The system can be used to make an automated teaching platform adaptable to the user, by managing eventual drops in the cognitive and emotional engagement. The effectiveness of the learning process mainly depends on the engagement level of the learner. In case of distraction, lack of interest or superficial participation, the teaching strategy could be personalized by an automatic modulation of contents and communication strategies. The system is validated by an experimental case study on twenty-one students. The experimental task was to learn how a specific human-machine interface works. Both the cognitive and motor skills of participants were involved. De facto standard stimuli, namely (1) cognitive task (Continuous Performance Test), (2) music background (Music Emotion Recognition-MER database), and (3) social feedback (Hermans and De Houwer database), were employed to guarantee a metrologically founded reference. In withinsubject approach, the proposed signal processing pipeline (Filter bank, Common Spatial Pattern, and Support Vector Machine), reaches almost 77% average accuracy, in detecting both cognitive and emotional engagement. Man's relationship with knowledge is increasingly mediated by technology. Since the second half of the last century, the digital era, namely the period of the pervasive use of information and communication technologies in every area of life, has had a major impact on the human learning 1. Currently, the ongoing Fourth Industrial Revolution (Industry 4.0) further expands the role of technology in learning processes: automated teaching platforms can adapt in real-time to the user skills and the new generation interfaces allow multi-sensorial interactions with virtual contents 2-4. In the pedagogical domain, the concept of "Learning 4.0" is emerging and it is not just a marketing gimmick 5. The 4.0 technologies are strongly impacting on the creation, the conservation, and the transmission of knowledge 6. In particular, the new immersive eXtended Reality (XR) solutions make possible to achieve embodied learning by enhancing the catalytic learning role of bodily activities 7. Furthermore, wearable transducers and embedded Artificial Intelligence (AI) increase real-time adaptivity in Human-Machine Interaction 8. In detail, in the Learning 4.0 context, the adaptation between humans and machines is reciprocal: the subject learns to use the human-machine interface, but also the machine adapts to human by learning from her/him 9. Traditionally, learning how to use a new technology interface was a once-in-a-lifetime effort conducted at a young age. For many people this has occurred with learning to read and write. Recently, the rapidity of technological evolution has been entailing the need to learn how to use several interfaces. The joy-pad, icon, touch/ multi-touch screen, speech and gesture recognition are examples of the evolution of new interfaces (hardware and software components). More specifically, learning to use an interface is a hard task which requires complex cognitive-motor skills. When human beings learned to use the mouse/touchscreen, as well as when they learned to write, read or speak, their minds learned complex cognitive-body patterns 10,11. Regarding the human-machine interfaces of older generations, the user was autonomously required to explore the different available resources and learn their use. Currently, the Interfaces 4.0 can adapt in real time to the use, by supporting the learning process 2,3,12. Proper adaptive strategies could be aimed to improve the learner engagement. Indeed, according to the literature, the effectiveness learning process depends on the engagement level of the learner. In a study on the role of learning
Machine Learning Strategies to Improve Cross-Subject and Cross-Session Generalization in EEG-Based Emotion Recognition: A Systematic Review
arXiv (Cornell University), Oct 12, 2022
1 An interesting case of the well-known Dataset Shift Problem is the classification of Electroenc... more 1 An interesting case of the well-known Dataset Shift Problem is the classification of Electroencephalogram (EEG) signals in the context of Brain-Computer Interface (BCI). The non-stationarity of EEG signals can lead to poor generalisation performance in BCI classification systems used in different sessions, also from the same subject. In this paper, we start from the hypothesis that the Dataset Shift problem can be alleviated by exploiting suitable eXplainable Artificial Intelligence (XAI) methods to locate and transform the relevant characteristics of the input for the goal of classification. In particular, we focus on an experimental analysis of explanations produced by several XAI methods on an ML system trained on a typical EEG dataset for emotion recognition. Results show that many relevant components found by XAI methods are shared across the sessions and can be used to build a system able to generalise better. However, relevant components of the input signal also appear to be highly dependent on the input itself.
Adoption of Machine Learning Techniques to Enhance Classification Performance in Reactive Brain-Computer Interfaces
2022 IEEE International Symposium on Medical Measurements and Applications (MeMeA), Jun 22, 2022
Engineering Applications of Artificial Intelligence, Aug 1, 2023
In the Machine Learning (ML) literature, a well-known problem is the Dataset Shift problem where,... more In the Machine Learning (ML) literature, a well-known problem is the Dataset Shift problem where, differently from the ML standard hypothesis, the data in the training and test sets can follow different probability distributions, leading ML systems toward poor generalisation performances. This problem is intensely felt in the Brain-Computer Interface (BCI) context, where bio-signals as Electroencephalographic (EEG) are often used. In fact, EEG signals are highly non-stationary both over time and between different subjects. To overcome this problem, several proposed solutions are based on recent transfer learning approaches such as Domain Adaption (DA). In several cases, however, the actual causes of the improvements remain ambiguous. This paper focuses on the impact of data normalisation, or standardisation strategies applied together with DA methods. In particular, using SEED, DEAP, and BCI Competition IV 2a EEG datasets, we experimentally evaluated the impact of different normalization strategies applied with and without several well-known DA methods, comparing the obtained performances. It results that the choice of the normalisation strategy plays a key role on the classifier performances in DA scenarios, and interestingly, in several cases, the use of only an appropriate normalisation schema outperforms the DA technique.
Reproducible Assessment of Valence and Arousal Based on an EEG Wearable Device
2022 IEEE International Conference on Metrology for Extended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE), Oct 26, 2022
arXiv (Cornell University), Oct 3, 2022
In the Machine Learning (ML) literature, a well-known problem is the Dataset Shift problem where,... more In the Machine Learning (ML) literature, a well-known problem is the Dataset Shift problem where, differently from the ML standard hypothesis, the data in the training and test sets can follow different probability distributions, leading ML systems toward poor generalisation performances. This problem is intensely felt in the Brain-Computer Interface (BCI) context, where bio-signals as Electroencephalographic (EEG) are often used. In fact, EEG signals are highly non-stationary both over time and between different subjects. To overcome this problem, several proposed solutions are based on recent transfer learning approaches such as Domain Adaption (DA). In several cases, however, the actual causes of the improvements remain ambiguous. This paper focuses on the impact of data normalisation, or standardisation strategies applied together with DA methods. In particular, using SEED, DEAP, and BCI Competition IV 2a EEG datasets, we experimentally evaluated the impact of different normalization strategies applied with and without several well-known DA methods, comparing the obtained performances. It results that the choice of the normalisation strategy plays a key role on the classifier performances in DA scenarios, and interestingly, in several cases, the use of only an appropriate normalisation schema outperforms the DA technique.
arXiv (Cornell University), Dec 16, 2022
A systematic review on machine-learning strategies for improving generalizability (cross-subjects... more A systematic review on machine-learning strategies for improving generalizability (cross-subjects and cross-sessions) electroencephalography (EEG) based in emotion classification was realized. In this context, the nonstationarity of EEG signals is a critical issue and can lead to the Dataset Shift problem. Several architectures and methods have been proposed to * This work has been submitted for possible publication on a journal. Copyright may be transferred without notice.
arXiv (Cornell University), Jun 9, 2023
Explainable Artificial Intelligence (XAI) aims to provide insights into the decision-making proce... more Explainable Artificial Intelligence (XAI) aims to provide insights into the decision-making process of AI models, allowing users to understand their results beyond their decisions. A significant goal of XAI is to improve the performance of AI models by providing explanations for their decision-making processes. However, most XAI literature focuses on how to explain an AI system, while less attention has been given to how XAI methods can be exploited to improve an AI system. In this work, a set of well-known XAI methods typically used with Machine Learning (ML) classification tasks are investigated to verify if they can be exploited, not just to provide explanations but also to improve the performance of the model itself. To this aim, two strategies to use the explanation to improve a classification system are reported and empirically evaluated on three datasets: Fashion-MNIST, CIFAR10, and STL10. Results suggest that explanations built by Integrated Gradients highlight input features that can be effectively used to improve classification performance.
Acta IMEKO
An electroencephalography (EEG)-based classification system of three levels of fear of heights is... more An electroencephalography (EEG)-based classification system of three levels of fear of heights is proposed. A virtual reality (VR) scenario representing a canyon was exploited to gradually expose the subjects to fear inducing stimuli with increasing intensity. An elevating platform allowed the subjects to reach three different height levels. Psychometric tools were employed to initially assess the severity of fear of heights and to assess the effectiveness of fear induction. A feasibility study was conducted on eight subjects who underwent three experimental sessions. The EEG signals were acquired through a 32-channel headset during the exposure to the eliciting VR scenario. The main EEG bands and scalp regions were explored in order to identify which are the most affected by the fear of heights. As a result, the gamma band, followed by the high-beta band, and the frontal area of the scalp resulted the most significant. The average accuracies in the within-subject case for the three...
IEEE Access
This work involved human subjects or animals in its research. Approval of all ethical and experim... more This work involved human subjects or animals in its research. Approval of all ethical and experimental procedures and protocols was granted by the Ethical Committee of Federico II University.
Reproducible Assessment of Valence and Arousal Based on an EEG Wearable Device
2022 IEEE International Conference on Metrology for Extended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE)
EEG-based system for Executive Function fatigue detection
2022 IEEE International Conference on Metrology for Extended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE)
Hidden classification layers: Enhancing linear separability between classes in neural networks layers
Pattern Recognition Letters
IEEE Access
This work addresses the employment of Machine Learning (ML) and Domain Adaptation (DA) in the fra... more This work addresses the employment of Machine Learning (ML) and Domain Adaptation (DA) in the framework of Brain-Computer Interfaces (BCIs) based on Steady-State Visually Evoked Potentials (SSVEPs). Currently, all the state-of-the-art classification strategies do not consider the high non-stationarity typical of brain signals. This can lead to poor performance, expecially when short-time signals have to be considered to allow real-time human-environment interaction. In this regard, ML and DA techniques can represent a suitable strategy to enhance the performance of SSVEPs classification pipelines. In particular, the employment of a two-step DA technique is proposed: first, the standardization of the data per subject is performed by exploiting a part of unlabeled test data during the training stage; second, a similarity measure between subjects is considered in the selection of the validation sets. The proposal was applied to three classifiers to verify the statistical significance of the improvements over the standard approaches. These classifiers were validated and comparatively tested on a well-known public benchmark dataset. An appropriate validation method was used in order to simulate real-world usage. The experimental results show that the proposed approach significantly improves the classification accuracy of SSVEPs. In fact, up to 62.27 % accuracy was achieved also in the case of short-time signals (i.e., 1.0 s). This represents a further confirmation of the suitability of advanced ML to improve the performance of BCIs for daily-life applications.
Instrumentation for EEG-based monitoring of the executive functions in a dual-task framework
Proceedings of the 25th IMEKO TC4 International Symposium and 23rd International Workshop on ADC and DAC Modelling and Testing
IEEE Access
The market uptake of Brain-Computer Interface technologies for clinical and non-clinical applicat... more The market uptake of Brain-Computer Interface technologies for clinical and non-clinical applications is attracting the scientific world towards the development of daily-life wearable systems. Beyond the use of dry electrodes and wireless technology, reducing the number of channels is crucial to enhance the ergonomics of devices. This paper presents a review of the studies exploiting a number of channels less than 16 for electroencephalographic (EEG) based-emotion recognition. The main findings of this review concern: (i) the criteria to select the most promising scalp areas for EEG acquisitions; (ii) the attention to prior neurophysiological knowledge; and (iii) the convergences among different studies with respect to preferable areas of the scalp for signal acquisition. Three main approaches emerge for channel selection: datadriven, prior knowledge-based, and based on commercially-available wearable solutions. The most spread is the data-driven, but the neurophysiology of emotions is rarely taken into account. Furthermore, commercial EEG devices usually do not provide electrodes purposefully chosen to assess emotions. Considerable convergences emerge for some electrodes: Fp1, Fp2, F3 and F4 resulted the most informative channels for the valence dimension, according to both data-driven and neurophysiological prior knowledge approaches. The P3 and P4 resulted in being significant for the arousal dimension. INDEX TERMS Emotion, EEG, channel selection, machine learning, neurophysiology of emotions, wearable devices.
A central issue addressed by the rapidly growing research area of eXplainable Artificial Intellig... more A central issue addressed by the rapidly growing research area of eXplainable Artificial Intelligence (XAI) is to provide methods to give explanations for the behaviours of Machine Learning (ML) non-interpretable models after the training. Recently, it is becoming more and more evident that new directions to create better explanations should take into account what a good explanation is to a human user. This paper suggests taking advantage of developing an XAI framework that allows producing multiple explanations for the response of image a classification system in terms of potentially different middle-level input features. To this end, we propose an XAI framework able to construct explanations in terms of input features extracted by autoencoders. We start from the hypothesis that some autoencoders, relying on standard data representation approaches, could extract more salient and understandable input properties, which we call here Middle-Level input Features (MLFs), for a user with ...
Cornell University - arXiv, May 21, 2021
Over the last few years, we have witnessed the availability of an increasing data generated from ... more Over the last few years, we have witnessed the availability of an increasing data generated from non-Euclidean domains, which are usually represented as graphs with complex relationships, and Graph Neural Networks (GNN) have gained a high interest because of their potential in processing graphstructured data. In particular, there is a strong interest in exploring the possibilities in performing convolution on graphs using an extension of the GNN architecture, generally referred to as Graph Convolutional Neural Networks (ConvGNN). Convolution on graphs has been achieved mainly in two forms: spectral and spatial convolutions. Due to the higher flexibility in exploring and exploiting the graph structure of data, there is recently an increasing interest in investigating the possibilities that the spatial approach can offer. The idea of finding a way to adapt the network behaviour to the inputs they process to maximize the total performances has aroused much interest in the neural networks literature over the years. This paper presents a novel method to adapt the behaviour of a ConvGNN to the input proposing a method to perform spatial convolution on graphs using input-specific filters, which are dynamically generated from nodes feature vectors. The experimental assessment confirms the capabilities of the proposed approach, which achieves satisfying results using a low number of filters.
Uploads
Papers by Andrea Apicella