Zobrazeno 1 - 10
of 25
pro vyhledávání: '"Choudhury, Anamitra R."'
Autor:
Chakaravarthy, Venkatesan T., Seshadri, Padmanabha V., Aggarwal, Pooja, Choudhury, Anamitra R., Kumar, Ashok Pon, Sabharwal, Yogish, Singhee, Amith
In conventional public clouds, designing a suitable initial cluster for a given application workload is important in reducing the computational foot-print during run-time. In edge or on-premise clouds, cold-start rightsizing the cluster at the time o
Externí odkaz:
http://arxiv.org/abs/2112.11597
Autor:
Goyal, Saurabh, Choudhury, Anamitra R., Raje, Saurabh M., Chakaravarthy, Venkatesan T., Sabharwal, Yogish, Verma, Ashish
We develop a novel method, called PoWER-BERT, for improving the inference time of the popular BERT model, while maintaining the accuracy. It works by: a) exploiting redundancy pertaining to word-vectors (intermediate encoder outputs) and eliminating
Externí odkaz:
http://arxiv.org/abs/2001.08950
Autor:
Vooturi, Dharma Teja, Goyal, Saurabh, Choudhury, Anamitra R., Sabharwal, Yogish, Verma, Ashish
Large number of weights in deep neural networks makes the models difficult to be deployed in low memory environments such as, mobile phones, IOT edge devices as well as "inferencing as a service" environments on cloud. Prior work has considered reduc
Externí odkaz:
http://arxiv.org/abs/1711.00244
Publikováno v:
In Discrete Optimization November 2020 38
Autor:
Agarwal, Archita, Chakaravarthy, Venkatesan T., Choudhury, Anamitra R., Roy, Sambuddha, Sabharwal, Yogish
In this paper, we study a class of set cover problems that satisfy a special property which we call the {\em small neighborhood cover} property. This class encompasses several well-studied problems including vertex cover, interval cover, bag interval
Externí odkaz:
http://arxiv.org/abs/1312.7217
We describe a primal-dual framework for the design and analysis of online convex optimization algorithms for {\em drifting regret}. Existing literature shows (nearly) optimal drifting regret bounds only for the $\ell_2$ and the $\ell_1$-norms. Our wo
Externí odkaz:
http://arxiv.org/abs/1309.5904
Autor:
Saurabh Gupta, Choudhury, Anamitra R., Raje, Saurabh M., Chakaravarthy, Venkatesan T., Sabharwal, Yogish, Verma, Ashish
Publikováno v:
Web of Science
We develop a novel method, called PoWER-BERT, for improving the inference time of the popular BERT model, while maintaining the accuracy. It works by: a) exploiting redundancy pertaining to word-vectors (intermediate encoder outputs) and eliminating
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::103c9343bff40488551cfe811c081faa
Publikováno v:
Web of Science
Large number of weights in deep neural networks makes the models difficult to be deployed in low memory environments such as, mobile phones, IOT edge devices as well as "inferencing as a service" environments on cloud. Prior work has considered reduc
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d3b518ff89e4c079b08364766988ab37
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.