Zobrazeno 1 - 10
of 2 978
pro vyhledávání: '"Canini A."'
Progressing beyond centralized AI is of paramount importance, yet, distributed AI solutions, in particular various federated learning (FL) algorithms, are often not comprehensively assessed, which prevents the research community from identifying the
Externí odkaz:
http://arxiv.org/abs/2407.14154
This work tackles the challenges of data heterogeneity and communication limitations in decentralized federated learning. We focus on creating a collaboration graph that guides each client in selecting suitable collaborators for training personalized
Externí odkaz:
http://arxiv.org/abs/2406.06520
We propose NeuronaBox, a flexible, user-friendly, and high-fidelity approach to emulate DNN training workloads. We argue that to accurately observe performance, it is possible to execute the training workload on a subset of real nodes and emulate the
Externí odkaz:
http://arxiv.org/abs/2405.02969
Autor:
Alballa, Norah, Canini, Marco
This research investigates the enhancement of knowledge distillation (KD) processes in pre-trained models, an emerging field in knowledge transfer with significant implications for distributed training and federated learning environments. These envir
Externí odkaz:
http://arxiv.org/abs/2402.14922
In Federated Learning (FL), forgetting, or the loss of knowledge across rounds, hampers algorithm convergence, particularly in the presence of severe data heterogeneity among clients. This study explores the nuances of this issue, emphasizing the cri
Externí odkaz:
http://arxiv.org/abs/2402.05558
In distributed training, communication often emerges as a bottleneck. In response, we introduce Kimad, a solution that offers adaptive gradient compression. By consistently monitoring bandwidth, Kimad refines compression ratios to match specific neur
Externí odkaz:
http://arxiv.org/abs/2312.08053
Efficient distributed training is a principal driver of recent advances in deep learning. However, communication often proves costly and becomes the primary bottleneck in these systems. As a result, there is a demand for the design of efficient commu
Externí odkaz:
http://arxiv.org/abs/2305.18627
Federated learning, an emerging machine learning paradigm, enables clients to collaboratively train a model without exchanging local data. Clients participating in the training process significantly impact the convergence rate, learning efficiency, a
Externí odkaz:
http://arxiv.org/abs/2302.06599
Autor:
Ivanov, Andrei, Rothenberger, Benjamin, Dethise, Arnaud, Canini, Marco, Hoefler, Torsten, Perrig, Adrian
With the application of machine learning to security-critical and sensitive domains, there is a growing need for integrity and privacy in computation using accelerators, such as GPUs. Unfortunately, the support for trusted execution on GPUs is curren
Externí odkaz:
http://arxiv.org/abs/2209.03125
Antarctic rock and soil microbiomes: Shared taxa, selective pressures, and extracellular DNA effects
Publikováno v:
Geoderma, Vol 446, Iss , Pp 116918- (2024)
Highly adapted and often endemic microbial taxa inhabit soils and rocks of extremely cold and dry Antarctic deserts. However, the source populations of these organisms have not yet been clarified. Local hotspots, rather than worldwide wind dispersion
Externí odkaz:
https://doaj.org/article/46a75d42fb6048b4aa7ac7a196576928