École Polytechnique Fédérale de Lausanne

About: École Polytechnique Fédérale de Lausanne is a facility organization based out in Lausanne, Switzerland. It is known for research contribution in the topics: Population & Catalysis. The organization has 44041 authors who have published 98296 publications receiving 4372092 citations. The organization is also known as: EPFL & ETHL.

An introduction to the multi-user MIMO downlink

Abstract: Multiple-input multiple-output (MIMO) communication techniques have been an important area of focus for next-generation wireless systems because of their potential for high capacity, increased diversity, and interference suppression. For applications such as wireless LANs and cellular telephony, MIMO systems will likely be deployed in environments where a single base must communicate with many users simultaneously. As a result, the study of multi-user MIMO systems has emerged recently as an important research topic. Such systems have the potential to combine the high capacity achievable with MIMO processing with the benefits of space-division multiple access. In this article we review several algorithms that have been proposed with this goal in mind. We describe two classes of solutions. The first uses a signal processing approach with various types of transmitter beamforming. The second uses "dirty paper" coding to overcome the interference a user sees from signals intended for other users. We conclude by describing future areas of research in multi-user MIMO communications.

VarSome: the human genomic variant search engine

Abstract: Summary VarSome.com is a search engine, aggregator and impact analysis tool for human genetic variation and a community-driven project aiming at sharing global expertise on human variants. Availability and implementation VarSome is freely available at http://varsome.com. Supplementary information Supplementary data are available at Bioinformatics online.

Data-Driven Distributionally Robust Optimization Using the Wasserstein Metric: Performance Guarantees and Tractable Reformulations

Abstract: We consider stochastic programs where the distribution of the uncertain parameters is only observable through a finite training dataset. Using the Wasserstein metric, we construct a ball in the space of (multivariate and non-discrete) probability distributions centered at the uniform distribution on the training samples, and we seek decisions that perform best in view of the worst-case distribution within this Wasserstein ball. The state-of-the-art methods for solving the resulting distributionally robust optimization problems rely on global optimization techniques, which quickly become computationally excruciating. In this paper we demonstrate that, under mild assumptions, the distributionally robust optimization problems over Wasserstein balls can in fact be reformulated as finite convex programs—in many interesting cases even as tractable linear programs. Leveraging recent measure concentration results, we also show that their solutions enjoy powerful finite-sample performance guarantees. Our theoretical results are exemplified in mean-risk portfolio optimization as well as uncertainty quantification.

SCAFFOLD: Stochastic Controlled Averaging for Federated Learning

Abstract: Federated Averaging (FedAvg) has emerged as the algorithm of choice for federated learning due to its simplicity and low communication cost. However, in spite of recent research efforts, its performance is not fully understood. We obtain tight convergence rates for FedAvg and prove that it suffers from `client-drift' when the data is heterogeneous (non-iid), resulting in unstable and slow convergence. As a solution, we propose a new algorithm (SCAFFOLD) which uses control variates (variance reduction) to correct for the `client-drift' in its local updates. We prove that SCAFFOLD requires significantly fewer communication rounds and is not affected by data heterogeneity or client sampling. Further, we show that (for quadratics) SCAFFOLD can take advantage of similarity in the client's data yielding even faster convergence. The latter is the first result to quantify the usefulness of local-steps in distributed optimization.

A nanoporous molybdenum carbide nanowire as an electrocatalyst for hydrogen evolution reaction

Abstract: A highly active and stable electrochemical catalyst of nanoporous molybdenum carbide nanowires (np-Mo2C NWs) has been developed for hydrogen evolution reaction (HER). The np-Mo2C NWs were synthesized simply by pyrolysis of a MoOx/amine hybrid precursor with sub-nanosized periodic structure under an inert atmosphere. The enriched nanoporosity and large reactive surface of these highly dispersed nanowires with uniform Mo2C nanocrystallites provide an efficient electrocatalysis, leading to their superior HER activity with lower onset overpotential and higher current densities than Mo2C microparticles. This study opens a new perspective for the development of highly active non-noble electrocatalysts for hydrogen production from water splitting.