ABSTRACT
Spectral preprocessing techniques can, to a certain extent, eliminate irrelevant information, such as current noise and stray light from spectral data, thereby enhancing the performance of prediction models. However, current preprocessing techniques mostly attempt to find the best single preprocessing method or their combination, overlooking the complementary information among different preprocessing methods. These preprocessing techniques fail to maximize the utilization of useful information in spectral data and restrict the performance of prediction models. This study proposed a spectral ensemble preprocessing method based on the rapidly developing ensemble learning methods in recent years and the ridge regression (RR) model, named stacking preprocessing ridge regression (SPRR), to address the aforementioned issues. Different from conventional ensemble learning methods, the proposed SPRR method applied multiple different preprocessing techniques to the original spectral data, generating multiple preprocessed datasets. These datasets were then individually inputted into RR base models for training. Ultimately, RR still served as the meta-model, integrating the output results of each RR base model through stacking. This approach not only produced diversity in base models but also achieved higher accuracy and lower computational complexity by using a single type of base model. On the apple spectral dataset collected by our team, correlation analysis showed significant complementary information among the data produced by different preprocessing techniques. This provided robust theoretical support for the proposed SPRR method. By introducing the currently popular averaging ensemble preprocessing method in a comparative experiment, the results of applying the proposed SPRR method to six datasets (apple, meat, wheat, olive oil, tablet, and corn) demonstrated that compared to the single preprocessing method and averaging ensemble preprocessing method, SPRR yielded the best accuracy and reliability for all six datasets. Furthermore, under the same conditions of the training and test datasets, the proposed SPRR method demonstrated better performance than the four commonly used ensemble preprocessing methods.
ABSTRACT
A new information theoretic condition is presented for reconstructing a discrete random variable X based on the knowledge of a set of discrete functions of X. The reconstruction condition is derived from Shannon's 1953 lattice theory with two entropic metrics of Shannon and Rajski. Because such a theoretical material is relatively unknown and appears quite dispersed in different references, we first provide a synthetic description (with complete proofs) of its concepts, such as total, common, and complementary information. The definitions and properties of the two entropic metrics are also fully detailed and shown to be compatible with the lattice structure. A new geometric interpretation of such a lattice structure is then investigated, which leads to a necessary (and sometimes sufficient) condition for reconstructing the discrete random variable X given a set {X1, ,Xn} of elements in the lattice generated by X. Intuitively, the components X1, ,Xn of the original source of information X should not be globally "too far away" from X in the entropic distance in order that X is reconstructable. In other words, these components should not overall have too low of a dependence on X; otherwise, reconstruction is impossible. These geometric considerations constitute a starting point for a possible novel "perfect reconstruction theory", which needs to be further investigated and improved along these lines. Finally, this condition is illustrated in five specific examples of perfect reconstruction problems: the reconstruction of a symmetric random variable from the knowledge of its sign and absolute value, the reconstruction of a word from a set of linear combinations, the reconstruction of an integer from its prime signature (fundamental theorem of arithmetic) and from its remainders modulo a set of coprime integers (Chinese remainder theorem), and the reconstruction of the sorting permutation of a list from a minimal set of pairwise comparisons.
ABSTRACT
Accurate skin lesion diagnosis requires a great effort from experts to identify the characteristics from clinical and dermoscopic images. Deep multimodal learning-based methods can reduce intra- and inter-reader variability and improve diagnostic accuracy compared to the single modality-based methods. This study develops a novel method, named adversarial multimodal fusion with attention mechanism (AMFAM), to perform multimodal skin lesion classification. Specifically, we adopt a discriminator that uses adversarial learning to enforce the feature extractor to learn the correlated information explicitly. Moreover, we design an attention-based reconstruction strategy to encourage the feature extractor to concentrate on learning the features of the lesion area, thus, enhancing the feature vector from each modality with more discriminative information. Unlike existing multimodal-based approaches, which only focus on learning complementary features from dermoscopic and clinical images, our method considers both correlated and complementary information of the two modalities for multimodal fusion. To verify the effectiveness of our method, we conduct comprehensive experiments on a publicly available multimodal and multi-task skin lesion classification dataset: 7-point criteria evaluation database. The experimental results demonstrate that our proposed method outperforms the current state-of-the-art methods and improves the average AUC score by above 2% on the test set.
Subject(s)
Diagnostic Imaging , Skin Diseases , Skin , Databases, Factual , Humans , Machine Learning , Skin/pathology , Skin Diseases/classification , Skin Diseases/diagnosisABSTRACT
What are the distinct ways in which a set of predictor variables can provide information about a target variable? When does a variable provide unique information, when do variables share redundant information, and when do variables combine synergistically to provide complementary information? The redundancy lattice from the partial information decomposition of Williams and Beer provided a promising glimpse at the answer to these questions. However, this structure was constructed using a much criticised measure of redundant information, and despite sustained research, no completely satisfactory replacement measure has been proposed. In this paper, we take a different approach, applying the axiomatic derivation of the redundancy lattice to a single realisation from a set of discrete variables. To overcome the difficulty associated with signed pointwise mutual information, we apply this decomposition separately to the unsigned entropic components of pointwise mutual information which we refer to as the specificity and ambiguity. This yields a separate redundancy lattice for each component. Then based upon an operational interpretation of redundancy, we define measures of redundant specificity and ambiguity enabling us to evaluate the partial information atoms in each lattice. These atoms can be recombined to yield the sought-after multivariate information decomposition. We apply this framework to canonical examples from the literature and discuss the results and the various properties of the decomposition. In particular, the pointwise decomposition using specificity and ambiguity satisfies a chain rule over target variables, which provides new insights into the so-called two-bit-copy example.
ABSTRACT
The formulation of the Partial Information Decomposition (PID) framework by Williams and Beer in 2010 attracted a significant amount of attention to the problem of defining redundant (or shared), unique and synergistic (or complementary) components of mutual information that a set of source variables provides about a target. This attention resulted in a number of measures proposed to capture these concepts, theoretical investigations into such measures, and applications to empirical data (in particular to datasets from neuroscience). In this Special Issue on "Information Decomposition of Target Effects from Multi-Source Interactions" at Entropy, we have gathered current work on such information decomposition approaches from many of the leading research groups in the field. We begin our editorial by providing the reader with a review of previous information decomposition research, including an overview of the variety of measures proposed, how they have been interpreted and applied to empirical investigations. We then introduce the articles included in the special issue one by one, providing a similar categorisation of these articles into: i. proposals of new measures; ii. theoretical investigations into properties and interpretations of such approaches, and iii. applications of these measures in empirical studies. We finish by providing an outlook on the future of the field.
ABSTRACT
El estudio realizado indagó el papel que tiene la información complementaria en los procesos de resistencia que participan en el desarrollo de memorias colectivas. Se diseñó un experimento con el fin de evaluar si es posible fortalecer los procesos de resistencia proveyendo a los participantes con información complementaria al material estímulo. El experimento constaba de tres fases: El día 1 escucharon historias (cada sujeto escuchó una versión levemente modificada) y desarrollaron un recuerdo libre, individual y escrito acerca de cada una. El día 2 participaron en un recupero grupal acerca de las historias. El día 3 respondieron a tareas escritas de recuerdo libre y de reconocimiento forzado e indicaron el nivel de confianza en sus respuestas. La calidad de los recuerdos se manipuló al proveer dibujos contextualizantes generales y específicos para el material estímulo. Además se generó una situación de desconfianza en uno de los participantes a fines de evaluar los procesos resistenciales. Los resultados indican que la información complementaria reforzante no tiene una incidencia mayor que la información contextualizante general en los procesos de resistencia. Si la fuente que impone no es confiable y escuchó la información sin dibujo que la acompañe, aumenta el nivel de confianza que el sujeto tiene en su propia respuesta. En cambio, si la fuente recibió la información con un dibujo acompañante, el nivel de confianza en la respuesta dada se mantiene igual, tanto si la fuente es confiable como si no lo es.
This study focuses on the resistance processes involved in the development of collective memories. It is part of a series of studies that consider conversational dynamics as a factor in the development of these memories. Previous studies have shown how conversation participates in the development and the shaping of collective memories. Recall of the past has shown to be very structured according to different roles: those of Narrator, Mentor and Monitor. Narrators, those subjects in a group recall who do most of the telling of the past, proved to be very effective in imposing their version of the past onto other members of the group. This fact would explain one of the ways memory converge and how collective memories are formed. When a group of subjects recall together the past, some of the processes involved are those of resistance and appropriation. In conversational recalls, participants have resisted the rendering of the past of some members and appropriated those of others. There are several studies that consider how different factors -psychological as well as social- have an impact on the formation of collective memories through conversation. These studies have shown how the quality of the memories affects the levels of resistance, as well as the dynamics of the conversation. Another factor that has had an impact on resistance and appropriation processes is trust. Entrusted subjects can be resisted in a group recall under certain circumstances. Finally, discussions about disagreements of the past, as they happen during a conversational recall of the past, have also proven to be an important factor that has had an effect on resistance levels. In this experimental study we investigated how complementary information may play a role on resistance processes. The study was designed in order to find out whether it is possible to reinforce resistance by providing the subjects with information that complements the stimuli material. It took place on three consecutive days. The sample was composed of 80 participants that conformed 20 groups of four subjects each. On the first day, subjects listened to some stories (each one heard a slightly different version), then they were exposed to a distracter, and then responded to a free recall task. On the second day, subjects were assembled around a table and asked to recall the stories together. Before the recall, a warning was provided over one of the subjects, who was not aware of the situation. Finally, on the third day, they completed a free recall and a forced recognition task remembering the stories as they heard them on the first day. In order to evaluate their confidence on the response, subjects had also to indicate how confidence they were in a 1 to 6 Lickert scale. By providing the subjects with specific and general contextual drawings together with the stimuli material, we manipulated the quality of the memories. Subjects with a contextual drawing would be able to form better memories of the stories, thus, more resistance to imposition. Would a specific drawing generate even more resistant memories? The mistrust situation generated on day 2 would allow enquiring on resistance processes. The findings indicate that reinforcing complementary information has no major incidence on resistance processes than general contextualizing information. Results on recognition have shown that. On the other hand, confidence rating analysis show that if the source imposing its version is not trusted and heard the story without complementary information, subjects tend to trust more on their own answers. Also, if the source imposing its version was provided with complementary information, subjects' confidence on their own answers remains the same whether the source is trusted or not.
ABSTRACT
Benefits to speech intelligibility can be achieved by enhancing a listener's ability to decipher it. However, much remains to be learned about the variables that influence the effectiveness of various listener-based manipulations. This study examined the benefit of providing listeners with the topic of some phases produced by speakers with either hypokinetic or ataxic dysarthria. Total and topic word accuracy, topic-related substitutions, and lexical boundary errors were calculated from the listener transcripts. Data were compared with those who underwent a familiarization process (reported by Liss, Spitzer, Caviness, & Adler, 2002) and with those inexperienced with disordered speech (reported by Liss Spitzer, Caviness, & Adler, 2000). Results revealed that listeners of ataxic speech provided with topic knowledge obtained higher intelligibility scores than naïve listeners. The magnitude of benefit was similar to the familiarization condition. However, topic word and word substitution analyses revealed different underlying perceptual mechanisms responsible for the observed benefit. No differences attributable to listening condition were discovered in lexical segmentation patterns. Overall, the results support the need for further study of listener-based manipulations to elucidate the mechanisms responsible for the observed perceptual benefits for each dysarthria type.