Zobrazeno 1 - 9
of 9
pro vyhledávání: '"Daniele Castellana"'
Autor:
Daniele Castellana, Davide Bacciu
Publikováno v:
Neurocomputing
Learning machines for structured data (e.g., trees) are intrinsically based on their capacity to learn representations by aggregating information from the multi-way relationships emerging from the structure topology. While complex aggregation functio
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1272cfd1caaffa6765608b9d70423ae4
http://hdl.handle.net/11568/1126484
http://hdl.handle.net/11568/1126484
Autor:
Daniele Castellana, Davide Bacciu
Publikováno v:
Neurocomputing. 342:49-59
The paper deals with the problem of unsupervised learning with structured data, proposing a mixture model approach to cluster tree samples. First, we discuss how to use the Switching-Parent Hidden Tree Markov Model, a compositional model for learning
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d7ddd11228eed8e111f82d5335e18d1c
https://doi.org/10.1016/j.neucom.2018.11.091
https://doi.org/10.1016/j.neucom.2018.11.091
We propose a conceptual model comprising a cascade of tipping points as a mechanism for past abrupt climate changes. In the model, changes in a control parameter, which could for instance be related to changes in the atmospheric circulation, induce s
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d8d0b622ce135a36eb4e82d3726ca793
https://doi.org/10.5194/esd-2021-7
https://doi.org/10.5194/esd-2021-7
Autor:
Daniele Castellana
Each research question in climate science requires an appropriate climate model to be formulated, where specific approximations are made and a certain number of physical processes is considered. Conceptual climate models represent only fundamental pr
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1d33380eaada65582434a27b71762f5f
https://doi.org/10.33540/240
https://doi.org/10.33540/240
Autor:
Daniele Castellana, Davide Bacciu
Publikováno v:
IJCNN
Most machine learning models for structured data encode the structural knowledge of a node by leveraging simple aggregation functions (in neural models, typically a weighted sum) of the information in the node's neighbourhood. Nevertheless, the choic
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::24a3f1194e8235022aeb80b06ffedfd8
https://doi.org/10.1109/ijcnn48605.2020.9206597
https://doi.org/10.1109/ijcnn48605.2020.9206597
Autor:
Davide Bacciu, Daniele Castellana
Publikováno v:
COLING
Proceedings of the 28th International Conference on Computational Linguistics
Proceedings of the 28th International Conference on Computational Linguistics
Processing sentence constituency trees in binarised form is a common and popular approach in literature. However, constituency trees are non-binary by nature. The binarisation procedure changes deeply the structure, furthering constituents that inste
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0bfc5f9e7d0b4337c414780cea1b4fb9
https://doi.org/10.18653/v1/2020.coling-main.346
https://doi.org/10.18653/v1/2020.coling-main.346
Autor:
Daniele Castellana, Bacciu, D.
Publikováno v:
Daniele Castellana
The paper introduces two new aggregation functions to encode structural knowledge from tree-structured data. They leverage the Canonical and Tensor-Train decompositions to yield expressive context aggregation while limiting the number of model parame
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4100e592f78c966860eb3bc1fe8740c6
Autor:
Daniele Castellana, Davide Bacciu
Publikováno v:
IJCNN
Bottom-Up Hidden Tree Markov Model is a highly expressive model for tree-structured data. Unfortunately, it cannot be used in practice due to the intractable size of its state-transition matrix. We propose a new approximation which lies on the Tucker
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::be5b34641c63ce976a851625eb6cd485
https://doi.org/10.1109/ijcnn.2019.8851851
https://doi.org/10.1109/ijcnn.2019.8851851