Bargaining Compatible Explanations

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Within the framework of ensemble methods, we investigate on a compatible learning scheme, denoted as learning by gossip with the aim of assessing its feasibility when facing a rather complex target function. Compatibility is in terms of probability that the learned function could be actually at the basis of the observed training set, hence an explanation of it. Feasibility is in terms of the related MSE on test sets. We base or conclusions on both theoretical and numerical arguments that are tossed on a well known benchmark.

Original languageBritish English
Title of host publicationProceedings - 2019 IEEE International Conference on Cognitive Computing, ICCC 2019 - Part of the 2019 IEEE World Congress on Services
EditorsElisa Bertino, Carl K. Chang, Peter Chen, Ernesto Damiani, Michael Goul, Katsunori Oyama
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages98-105
Number of pages8
ISBN (Electronic)9781728127118
DOIs
StatePublished - Jul 2019
Event4th IEEE International Conference on Cognitive Computing, ICCC 2019 - Milan, Italy
Duration: 8 Jul 201913 Jul 2019

Publication series

NameProceedings - 2019 IEEE International Conference on Cognitive Computing, ICCC 2019 - Part of the 2019 IEEE World Congress on Services

Conference

Conference4th IEEE International Conference on Cognitive Computing, ICCC 2019
Country/TerritoryItaly
CityMilan
Period8/07/1913/07/19

Keywords

  • Compatible-explanation
  • Ensemble-learning
  • Learning-by-gossip
  • Subsymbolic-kernels

Fingerprint

Dive into the research topics of 'Bargaining Compatible Explanations'. Together they form a unique fingerprint.

Cite this