R
Róbert Ormándi
Researcher at University of Szeged
Publications - 27
Citations - 770
Róbert Ormándi is an academic researcher from University of Szeged. The author has contributed to research in topics: Stochastic gradient descent & Online machine learning. The author has an hindex of 10, co-authored 27 publications receiving 610 citations. Previous affiliations of Róbert Ormándi include Google & Hungarian Academy of Sciences.
Papers
More filters
Posted Content
Large scale distributed neural network training through online distillation
Rohan Anil,Gabriel Pereyra,Alexandre Passos,Róbert Ormándi,George E. Dahl,Geoffrey E. Hinton +5 more
TL;DR: This paper claims that online distillation is a cost-effective way to make the exact predictions of a model dramatically more reproducible and can still speed up training even after the authors have already reached the point at which additional parallelism provides no benefit for synchronous or asynchronous stochastic gradient descent.
Journal ArticleDOI
Gossip learning with linear models on fully distributed data
TL;DR: This work proposes gossip learning, a generic approach that is based on multiple models taking random walks over the network in parallel, while applying an online learning algorithm to improve themselves, and getting combined via ensemble learning methods.
Proceedings Article
Gossip-based distributed stochastic bandit algorithms
Balázs Szörényi,Balázs Szörényi,Róbert Busa-Fekete,Róbert Busa-Fekete,Istvan Hegedus,Róbert Ormándi,Márk Jelasity,Balázs Kégl +7 more
TL;DR: This work shows that the probability of playing a suboptimal arm at a peer in iteration t = Ω(log N) is proportional to 1/(Nt) where N denotes the number of peers participating in the network.
Proceedings Article
Large scale distributed neural network training through online distillation
Rohan Anil,Gabriel Pereyra,Alexandre Passos,Róbert Ormándi,George E. Dahl,Geoffrey E. Hinton +5 more
TL;DR: In this paper, the authors explore a variant of distillation which is relatively straightforward to use as it does not require a complicated multi-stage setup or many new hyperparameters.
Journal ArticleDOI
Semi-automated Construction of Decision Rules to Predict Morbidities from Clinical Texts
Richárd Farkas,György Szarvas,György Szarvas,István Hegedűs,Attila Almási,Veronika Vincze,Róbert Ormándi,Róbert Busa-Fekete,Róbert Busa-Fekete +8 more
TL;DR: The results demonstrate the feasibility of the authors approach and show that even very simple systems with a shallow linguistic analysis can achieve remarkable accuracy scores for classifying clinical records on a limited set of concepts.