M
Manas Joglekar
Researcher at Stanford University
Publications - 39
Citations - 1055
Manas Joglekar is an academic researcher from Stanford University. The author has contributed to research in topics: Tuple & Time complexity. The author has an hindex of 19, co-authored 39 publications receiving 891 citations. Previous affiliations of Manas Joglekar include Indian Institute of Technology Bombay & Microsoft.
Papers
More filters
Journal ArticleDOI
Challenges in Data Crowdsourcing
TL;DR: An overview of data crowdsourcing is provided, giving examples of problems that the authors have tackled, and presenting the key design steps involved in implementing a crowdsourced solution.
Proceedings ArticleDOI
Neural Input Search for Large Scale Recommendation Models
Manas Joglekar,Cong Li,Mei Chen,Taibai Xu,Xiaoming Wang,Jay K. Adams,Pranav Khaitan,Jiahui Liu,Quoc V. Le +8 more
TL;DR: Neural Input Search (NIS) as mentioned in this paper uses reinforcement learning to find the optimal vocabulary size for each feature and embedding dimension for each value of the feature during training, which can find significantly better models with much fewer embedding parameters.
Posted Content
Evaluating the Crowd with Confidence
TL;DR: In this article, the authors devise techniques to generate confidence intervals for worker error rate estimates, thereby enabling a better evaluation of worker quality, and demonstrate wide applicability by using them to evict poorly performing workers and provide confidence intervals on the accuracy of the answers.
Proceedings ArticleDOI
Evaluating the crowd with confidence
TL;DR: This work devise techniques to generate confidence intervals for worker error rate estimates, thereby enabling a better evaluation of worker quality, and demonstrate wide applicability by using them to evict poorly performing workers, and provide confidence intervals on the accuracy of the answers.
Proceedings ArticleDOI
Transaction processing on confidential data using cipherbase
TL;DR: A prototype of Cipherbase is presented that uses FPGAs to provide secure processing and the system engineering details implemented to achieve competitive performance for transactional workloads are described.