Výsledky vyhledávání - "Arunachalam, Meena"

Report

PIFS-Rec: Process-In-Fabric-Switch for Large-Scale Recommendation System Inferences

Autor: Huo, Pingyi, Devulapally, Anusha, Maruf, Hasan Al, Park, Minseo, Nair, Krishnakumar, Arunachalam, Meena, Akbulut, Gulsum Gudukbay, Kandemir, Mahmut Taylan, Narayanan, Vijaykrishnan

Deep Learning Recommendation Models (DLRMs) have become increasingly popular and prevalent in today's datacenters, consuming most of the AI inference cycles. The performance of DLRMs is heavily influenced by available bandwidth due to their large vec

Externí odkaz: http://arxiv.org/abs/2409.16633

Zobrazit plný text záznamu

Report

Early prediction of onset of sepsis in Clinical Setting

Autor: Mohammad, Fahim, Arunachalam, Lakshmi, Sadhu, Samanway, Aasman, Boudewijn, Garg, Shweta, Ahmed, Adil, Colman, Silvie, Arunachalam, Meena, Kulkarni, Sudhir, Mirhaji, Parsa

This study proposes the use of Machine Learning models to predict the early onset of sepsis using deidentified clinical data from Montefiore Medical Center in Bronx, NY, USA. A supervised learning approach was adopted, wherein an XGBoost model was tr

Externí odkaz: http://arxiv.org/abs/2402.03486

Zobrazit plný text záznamu

Report

Strategies for Optimizing End-to-End Artificial Intelligence Pipelines on Intel Xeon Processors

Autor: Arunachalam, Meena, Sanghavi, Vrushabh, Yao, Yi A, Zhou, Yi A, Wang, Lifeng A, Wen, Zongru, Ammbashankar, Niroop, Wang, Ning W, Mohammad, Fahim

Publikováno v: The Parallel Universe Magazine, Issue 48, 2022

End-to-end (E2E) artificial intelligence (AI) pipelines are composed of several stages including data preprocessing, data ingestion, defining and training the model, hyperparameter optimization, deployment, inference, postprocessing, followed by down

Externí odkaz: http://arxiv.org/abs/2211.00286

Zobrazit plný text záznamu