conferences.bib

@inproceedings{FraLamRig20-LOD-IC,
  author = {Fraccaroli, Michele
and Lamma, Evelina
and Riguzzi, Fabrizio},
  editor = {Nicosia, Giuseppe
and Ojha, Varun
and La Malfa, Emanuele
and Jansen, Giorgio
and Sciacca, Vincenzo
and Pardalos, Panos
and Giuffrida, Giovanni
and Umeton, Renato},
  title = {Automatic Setting of {DNN} Hyper-Parameters by Mixing {Bayesian Optimization} and Tuning Rules},
  booktitle = {Machine Learning, Optimization, and Data Science, 6th International Conference, LOD 2020, Siena, Italy, July 19–23, 2020, Revised Selected Papers, Part I},
  year = {2020},
  publisher = {Springer International Publishing},
  address = {Cham},
  pages = {477--488},
  abstract = {Deep learning techniques play an increasingly important role in industrial and research environments due to their outstanding results. However, the large number of hyper-parameters to be set may lead to errors if they are set manually. The state-of-the-art hyper-parameters tuning methods are grid search, random search, and Bayesian Optimization. The first two methods are expensive because they try, respectively, all possible combinations and random combinations of hyper-parameters. Bayesian Optimization, instead, builds a surrogate model of the objective function, quantifies the uncertainty in the surrogate using Gaussian Process Regression and uses an acquisition function to decide where to sample the new set of hyper-parameters. This work faces the field of Hyper-Parameters Optimization (HPO). The aim is to improve Bayesian Optimization applied to Deep Neural Networks. For this goal, we build a new algorithm for evaluating and analyzing the results of the network on the training and validation sets and use a set of tuning rules to add new hyper-parameters and/or to reduce the hyper-parameter search space to select a better combination.},
  isbn = {978-3-030-64583-0},
  doi = {10.1007/978-3-030-64583-0_43},
  note = {The final publication is available at Springer via \url{https://link.springer.com/chapter/10.1007/978-3-030-64583-0_43}},
  copyright = {Springer},
  pdf = {http://ml.unife.it/wp-content/uploads/Papers/FraLamRig-LOD20.pdf},
  series = {Lecture Notes in Computer Science},
  volume = {12565}
}
@inproceedings{FraMazBiz21-ICTS4eHealth-IC,
  author = {Michele Fraccaroli and Giulia Mazzucchelli and Alice Bizzarri},
  title = {Machine Learning Techniques for Extracting Relevant Features from Clinical
	Data for {COVID-19} Mortality Prediction},
  booktitle = {2021 Symposium on Computers and Communications (ISCC): 26th IEEE Symposium
	on Computers and Communications - Workshop on ICT Solutions for eHealth
	(ICTS4eHealth) (ICTS4eHealth2021)},
  address = {Athens, Greece},
  days = 4,
  month = sep,
  year = 2021,
  keywords = {Machine Learning; COVID-19; Health Informatics; Explanation; Predictive
	Analysis},
  abstract = {The role of Machine Learning (ML) in healthcare is based on
	the ability of a machine to analyse the huge amounts of
	data available for each patient, like age, medical history,
	overall health status, test results, etc. With ML algorithms it is
	possible to learn models from data for the early identification
	of pathologies, and their severity. Early identification is crucial
	to proceed as soon as possible with the necessary therapeutic
	actions. This work applies modern ML techniques to clinical data
	of either COVID-19 positive and COVID-19 negative patients
	with pulmonary complications, to learn mortality prediction
	models for both groups of patients, and compare results. We
	have focused on symbolic methods for building
	classifiers able to extract patterns from clinical data. This
	approach leads to predictive Artificial Intelligence (AI) systems
	working on medical data, and able to explain the reasons that
	lead the systems themselves to reach a certain conclusion.},
  pages = {1-7},
  doi = {10.1109/ISCC53001.2021.9631477}
}

This file was generated by bibtex2html 1.98.