927 research outputs found
Attention-map augmentation for hypercomplex breast cancer classification
No full textBreast cancer is the most widespread neoplasm among women and early detection of this disease is critical. Deep learning techniques have become of great interest to improve diagnostic performance. However, distinguishing between malignant and benign masses in whole mammograms poses a challenge, as they appear nearly identical to an untrained eye, and the region of interest (ROI) constitutes only a small fraction of the entire image. In this paper, we propose a framework, parameterized hypercomplex attention maps (PHAM), to overcome these problems. Specifically, we deploy an augmentation step based on computing attention maps. Then, the attention maps are used to condition the classification step by constructing a multi-dimensional input comprised of the original breast cancer image and the corresponding attention map. In this step, a parameterized hypercomplex neural network (PHNN) is employed to perform breast cancer classification. The framework offers two main advantages. First, attention maps provide critical information regarding the ROI and allow the neural model to concentrate on it. Second, the hypercomplex architecture has the ability to model local relations between input dimensions thanks to hypercomplex algebra rules, thus properly exploiting the information provided by the attention map. We demonstrate the efficacy of the proposed framework on both mammography images as well as histopathological ones. We surpass attention-based state-of-the-art networks and the real-valued counterpart of our approach. The code of our work is available at https://github.com/ispamm/AttentionBCS
Learning visual stimulus-evoked EEG manifold for neural image classification
No full textVisual neural decoding, namely the ability to interpret external visual stimuli from patterns of brain activity, is a challenging task in neuroscience research.
Recent studies have focused on characterizing patterns of activity across multiple neurons that can be described in terms of population-level features. In this study, we combine spatial, spectral, and temporal features to achieve neural manifold classification capable to characterize visual perception and to simulate the working memory activity in the human brain. We treat spatio-temporal and spectral information separately by means of custom deep learning architectures based on Riemann manifold and the two-dimensional EEG spectrogram representation. In addition, a CNN-based classification model is used to classify visual stimulus-evoked EEG signals while viewing the 11-class (i.e., all-black plus 0-9 digit images) MindBigData Visual MNIST dataset. The effectiveness of the proposed integration strategy is evaluated on the stimulus-evoked EEG signal classification task, achieving an overall accuracy of {86\%}, comparable to state-of-the-art benchmarks
A Florentine family in crisis: the Strozzi in the fifteenth century.
No full textPhDIn 1434 the Strozzi lineage had held a leading position in
Florentine society and government for at least one hundred and fifty
years, and was one of the largest and wealthiest of the city's
patrician lineages. The records of the catasto of 1427 and of the
scrutiny of 1433 are used to give a profile of the dominant social,
economic and political position of the Strozzi before the advent of
Medicean dominance. Their record of electoral success, and the
political and cultural leadership of influential and respected men
such as Palla di Nofri and Matteo di Simone, with other factors, put
the Strozzi amongst the greatest enemies of the victorious Medicean
regime of late 1434. The effects of political opposition and exile
on the lineage are examined both directly, through records of office-holding,
and indirectly through such indicators as marriage alliances
and household wealth. The two most prominent lines of the Strozzi
were exiled after 1434. Palla di Nofri's life and preoccupations in
his Paduan exile are examined, together with the lives of his sons;
none of these Strozzi ever returned to Florence, pursued as they were
by the enmity of the Medicean regime. The very different careers of
Filippo di Matteo and his brother Lorenzo are also examined: how they
succeeded in founding a lucrative bank in Naples, and in returning to
Florence to 'rebuild' (rifare) the position of the Strozzi lineage
there. The final decades of the century saw the Strozzi in an
economically more secure position, due substantially to the efforts
of Filippo. Except for a very small number of its members admitted
into the regime, most of the lineage is here shown to have remained
excluded from significant political office until after the fall of
the Medici regime in 1494
A Reproducible Analysis of Sequential Recommender Systems
No full textSequential Recommender Systems (SRSs) have emerged as a highly efficient approach to recommendation systems. By leveraging sequential data, SRSs can identify temporal patterns in user behaviour, significantly improving recommendation accuracy and relevance. Ensuring the reproducibility of these models is paramount for advancing research and facilitating comparisons between them. Existing works exhibit shortcomings in reproducibility and replicability of results, leading to inconsistent statements across papers. Our work fills these gaps by standardising data pre-processing and model implementations, providing a comprehensive code resource, including a framework for developing SRSs and establishing a foundation for consistent and reproducible experimentation. We conduct extensive experiments on several benchmark datasets, comparing various SRSs implemented in our resource. We challenge prevailing performance benchmarks, offering new insights into the SR domain. For instance, SASRec does not consistently outperform GRU4Rec. On the contrary, when the number of model parameters becomes substantial, SASRec starts to clearly dominate all the other SRSs. This discrepancy underscores the significant impact that experimental configuration has on the outcomes and the importance of setting it up to ensure precise and comprehensive results. Failure to do so can lead to significantly flawed conclusions, highlighting the need for rigorous experimental design and analysis in SRS research. Our code is available at https://github.com/federicosiciliano/easy_lightning
Investigating the Robustness of Sequential Recommender Systems Against Training Data Perturbations
No full textSequential Recommender Systems (SRSs) are widely employed to model user
behavior over time. However, their robustness in the face of perturbations in
training data remains a largely understudied yet critical issue. A fundamental
challenge emerges in previous studies aimed at assessing the robustness of
SRSs: the Rank-Biased Overlap (RBO) similarity is not particularly suited for
this task as it is designed for infinite rankings of items and thus shows
limitations in real-world scenarios. For instance, it fails to achieve a
perfect score of 1 for two identical finite-length rankings. To address this
challenge, we introduce a novel contribution: Finite Rank-Biased Overlap
(FRBO), an enhanced similarity tailored explicitly for finite rankings. This
innovation facilitates a more intuitive evaluation in practical settings. In
pursuit of our goal, we empirically investigate the impact of removing items at
different positions within a temporally ordered sequence. We evaluate two
distinct SRS models across multiple datasets, measuring their performance using
metrics such as Normalized Discounted Cumulative Gain (NDCG) and Rank List
Sensitivity. Our results demonstrate that removing items at the end of the
sequence has a statistically significant impact on performance, with NDCG
decreasing up to 60%. Conversely, removing items from the beginning or middle
has no significant effect. These findings underscore the criticality of the
position of perturbed items in the training data. As we spotlight the
vulnerabilities inherent in current SRSs, we fervently advocate for intensified
research efforts to fortify their robustness against adversarial perturbations
Vues présentes sur le celtique cisalpin
No full textPresent views on Cisalpine Celtic.
During the last twenty years, the knowledge of Lepontic and of Italian Gaulish has been notably enriched by very important finds, and also by numerous studies. The author proposes a synthesis of all new linguistical and cultural data provided by epigraphy in this particular resort.Dans les vingt dernières années la connaissance du lépontique et du gaulois d’Italie a été notablement enrichie par certaines trouvailles fort importantes ainsi que par de nombreuses études. Tableau synthétique des nouvelles données linguistiques et culturelles apportées par l’épigraphie de ce domaine.Motta Filippo. Vues présentes sur le celtique cisalpin. In: Etudes Celtiques, vol. 29, 1992. Actes du IXe congrès international d'études celtiques. Paris, 7-12 juillet 1991. Deuxième partie : Linguistique, littératures. pp. 311-318
Occhi pieni e mani vaganti. Movimenti, emozioni, astrazioni.
No full textFilippo FIMIANI, Occhi pieni e mani vaganti. Movimenti, emozioni, astrazioni, «Fata Morgana», 12/2010, pp. 147-164. ISSN 1970-5786.
ITALIANO. Una piccola sequenza da un episodio della serie televisiva Mad Men in cui la relazione sensoriale emozionale tra lo spettatore e un corpo in movimento, precisamente un corpo danzante, è l’oggetto di questo articolo, che intende così verificare la cosiddetta naturalizzazione della fenomenologia e del corpo vissuto operata dalle scienze cognitive applicate agli studi visuali. L’autore mostra anche i rapporti tra la cosiddetta simulazione incarnata e una memoria simbolica e iconografica profonda, e discute in particolare la polarità tra proiezione e introiezione, attività e passività nell’esperienza empatica delle immagini in movimento.A short filmic example from the AMC's award-winning series Mad Men in which the sensorial and emotional relationship between the spectator and a moving body, i.e., a dancing body, is explicitly put into play. The essay wants to deal with the so called naturalization of phenomenology and of the living body, trained by the cognitive sciences reading the visuality. The author shows the relationship between the so called ‘embodied Simulation’ and a symbolic and iconographic deep memory of the pictures, and he gives an in-depth attention to the polarity between projections and introjections, activity and passivity of the spectator’s body during the empathic experience of the moving pictures
Multiple political geographies
No full textThis chapter delves into the multifaceted realm of political geography within the context of the journal Political Geography. The author, who has held various editorial positions within the journal, navigates the intricate terrain of diversity and inclusion, exploring the challenges posed by linguistic and cultural biases inherent in academic publishing. The chapter critically examines efforts to mitigate barriers and broaden the scope of political geography, questioning the adequacy of current approaches. Through a reflective analysis, the author elucidates the complex dynamics of political geography as a subdiscipline, shedding light on both the impediments and possibilities it offers to researchers and students
Erratum to: Lung Tissue Damage Caused by Heat Accumulation from Adjacent Laser Application: Surgical Implications (Thoracic and Cardiovascular Surgeon (2014) DOI: 10.1055/s-0034-1389086)
No full textIt has been brought to the Publisher's attention that the author names "Filippo Lococo" and "Alfredo Cesario" were not listed correctly in the above-mentioned article, published on eFirst on September 5, 2014. DOI of the original article is DOI: 10.1055/s-0034-1389086. The names should be read as "Filippo Lococo" and "Alfredo Cesario" in place of "Lococo Filippo" and "Cesario Alfredo."
Fasti di Lodovico XIV. il Grande : esposti in versi in occasione dell'esser levato al sacro fonte il primogenito del marchese Filippo cavalier Sampieri in nome di S.M. Cristianissima.
No full textIllustrated with etchings, comprising a portrait of Louis XIV by Francesco Maria Francia, and 12 plates by Lodovico Mattioli.Each of the twelve cantos is by a different author. Names supplied in manuscript in table of contents, p. [1] at end.Dedication signed: Filippo Sampieri.Imprint from colophon.Signatures: pi⁴ A-O⁸ P⁴.Mode of access: Internet.Bound in old vellum; ink title and library shelf mark on spine
- …
