D
David Buffoni
Researcher at Pierre-and-Marie-Curie University
Publications - 13
Citations - 266
David Buffoni is an academic researcher from Pierre-and-Marie-Curie University. The author has contributed to research in topics: Ranking (information retrieval) & Learning to rank. The author has an hindex of 5, co-authored 12 publications receiving 249 citations. Previous affiliations of David Buffoni include University of Paris.
Papers
More filters
Proceedings ArticleDOI
Ranking with ordered weighted pairwise classification
TL;DR: This work proposes to optimize a larger class of loss functions for ranking, based on an ordered weighted average (OWA) (Yager, 1988) of the classification losses, and shows that OWA aggregates of margin-based classification losses have good generalization properties.
Proceedings Article
Learning Scoring Functions with Order-Preserving Losses and Standardized Supervision
TL;DR: This work extends to ranking problems a notion of order-preserving losses previously introduced for multiclass classification, and shows that these losses lead to consistent formulations with respect to a family of ranking evaluation metrics.
Proceedings ArticleDOI
Predicting Popularity and Adapting Replication of Internet Videos for High-Quality Delivery
TL;DR: This work introduces Hermes, an adaptive replication scheme based on accurate predictions about the popularity of Internet videos that reduces storage usage for replication by two orders of magnitude, and under heavy load conditions, it increases the average bitrate provision by roughly 90%.
Journal ArticleDOI
A Learning to Rank framework applied to text-image retrieval
TL;DR: A framework based on a Learning to Rank setting for a text-image retrieval task and the state-of-the-art OWPC algorithm, called OWPC, which performs better than a simple baseline.
Book ChapterDOI
Random Forest Based on Federated Learning for Intrusion Detection
TL;DR: In this article , a federated learning approach that independently trains data subsets on multiple clients and sends only the resulting models for aggregation to a server was proposed. But the results showed that the global RF on the server achieved higher accuracy than the maximum achieved with individual RFs on the clients in the case of two out of four datasets and it was very close to the maximum for the third dataset.