Urmăriți
Saurabh Goyal
Saurabh Goyal
IBM Research, New York, USA, Indian Institute of Technology Delhi, IITK
Adresă de e-mail confirmată pe ibm.com
Titlu
Citat de
Citat de
Anul
Resource-efficient machine learning in 2 kb ram for the internet of things
A Kumar, S Goyal, M Varma
International conference on machine learning, 1935-1944, 2017
2912017
Protonn: Compressed and accurate knn for resource-scarce devices
C Gupta, AS Suggala, A Goyal, HV Simhadri, B Paranjape, A Kumar, ...
International conference on machine learning, 1331-1340, 2017
2042017
Power-bert: Accelerating bert inference via progressive word-vector elimination
S Goyal, AR Choudhury, S Raje, V Chakaravarthy, Y Sabharwal, A Verma
International Conference on Machine Learning, 3690-3699, 2020
752020
Compression of deep neural networks by combining pruning and low rank decomposition
S Goyal, AR Choudhury, V Sharma
2019 IEEE International Parallel and Distributed Processing Symposium …, 2019
122019
Efficient inferencing of compressed deep neural networks
DT Vooturi, S Goyal, AR Choudhury, Y Sabharwal, A Verma
arXiv preprint arXiv:1711.00244, 2017
52017
Variable batch size across layers for efficient prediction on CNNs
AR Choudhury, S Goyal, Y Sabharwal, V Ashish
IEEE Cloud, 2020
42020
Dynamic Batch Sizing for Inferencing of Deep Neural Networks in Resource-Constrained Environments
AR Choudhury, S Goyal, Y Sabharwal, A Verma, DT Vooturi
US Patent App. 16/167,727, 2020
22020
ProtoNER: Few shot Incremental Learning for Named Entity Recognition using Prototypical Networks
R Kumar, S Goyal, A Verma, V Isahagian
Business Process Management, 2023
12023
INFORMATION EXTRACTION FROM DOCUMENTS CONTAINING HANDWRITTEN TEXT
S Goyal, C Finegan-Dollak, A VERMA
US Patent App. P202106696US01, 2023
2023
FEW SHOT INCREMENTAL LEARNING FOR NAMED ENTITY RECOGNITION
R Kumar, S Goyal, A VERMA
US Patent App. P202202567US01, 2023
2023
ACCELERATING INFERENCE OF TRANSFORMER-BASED MODELS
S Goyal, V Chakaravarthy, AR Choudhury, S Raje, Y Sabharwal, A Verma
US Patent App. P202/007,353, 2021
2021
ACCELERATING INFERENCE OF NEURAL NETWORK MODELS VIA DYNAMIC EARLY EXITS
SM Raje, S Goyal, AR Choudhury, Y Sabharwal, A VERMA
US Patent App. P202005664US01, 2021
2021
A Method for Ordering the Inputs of Train Decomposition of Weight Matrices in Deep Neural Networks
V Chakaravarthy, S Goyal, Choudhury, Anamitra R, S Raje, Y Sabharwal, ...
US Patent App. P201909336US01, 2020
2020
Efficient Training with Variable Batch Sizing on gradient checkpoint segments
S Goyal, A Roy Choudhury, A Verma, Y Sabharwal
US Patent App. P201/904,457, 2019
2019
Dataset dependent low rank decomposition of Convolution neural networks
S Goyal, AR Choudhary, Y Sabharwal, V Chakaravarthy, A Verma
US Patent App. P201/809,644, 2019
2019
Sistemul nu poate realiza operația în acest moment. Încercați din nou mai târziu.
Articole 1–15