854 citations
- IBM Research - BrazilBR9 papers
- IBM (United States)US4 papers
- The University of MelbourneAU3 papers
- Institut national de recherche en sciences et technologies du numériqueFR2 papers
- Western Sydney UniversityAU2 papers
- Beihang UniversityCN1 paper
- Beth Israel Deaconess Medical CenterUS1 paper
- Cardiff UniversityGB1 paper
- Cleveland State UniversityUS1 paper
- Cracow University of TechnologyPL1 paper
- École Normale Supérieure de LyonFR1 paper
- Eindhoven University of TechnologyNL1 paper
4 papers · 1 filter
Nonstationary Reinforcement Learning with Linear Function Approximation
Huozhi Zhou, Jinglin Chen, Lav R. Varshney +1
We consider reinforcement learning (RL) in episodic Markov decision processes (MDPs) with linear function approximation under drifting environment. Specifically, both the reward an…
Wasserstein Barycenter Model Ensembling
Pierre Dognin, Igor Melnyk, Youssef Mroueh +3
In this paper we propose to perform model ensembling in a multiclass or a multilabel learning setting using Wasserstein (W.) barycenters. Optimal transport metrics, such as the Was…
Using link and content over time for embedding generation in Dynamic Attributed Networks
Ana Paula Appel, Renato L. F. Cunha, Charu C. Aggarwal +1
In this work, we consider the problem of combining link, content and temporal analysis for community detection and prediction in evolving networks. Such temporal and content-rich n…
Bus Travel Time Predictions Using Additive Models
Matthias Kormaksson, Luciano Barbosa, Marcos R. Vieira +1
Many factors can affect the predictability of public bus services such as traffic, weather and local events. Other aspects, such as day of week or hour of day, may influence bus tr…