@techreport{TR-IC-21-04,
   number = {IC-21-04},
   author  =  {Carlos  Avelar  and  João  Vitor  Gonçalves and Silvana
                   Trindade and Nelson L. S. da Fonseca},
   title   =   {{Vertical   Federated   Learning   for   Emulation  of 
                   Business-to-Business Applications at the Edge}},
   month = {February},
   year = {2021},
   institution = {Institute of Computing, University of Campinas},
   note = {In English, 12 pages.
    \par\selectlanguage{english}\textbf{Abstract}
       Given  the  ever-increasing  constraints and concerns regarding
       data  privacy  and  sharing,  a  method  to train collaborative
       machine  learning  models  without  exposing  training data can
       become  a  major  part of the way that data science is done. In
       this  work,  we  illustrate  the concepts of Vertical Federated
       Learning,  along  with  a  practical implementation emulating a
       real scenario of collaborative training of a model. We evaluate
       the  cost  associated  with homomorphic encryption that enables
       Federated  Learning  approaches  and  show  results of an MNIST
       solving model using Vertical Federated Learning.
  }
}