8 Papers
5 Citations
Kumar Abhishek is an academic researcher from International Institute of Information Technology, Hyderabad. The author has contributed to research in topics: Computer science & Common value auction. The author has an hindex of 1, co-authored 5 publications. Previous affiliations of Kumar Abhishek include International Institute of Information Technology.
Chat about Author
Papers
Towards improving e-commerce customer review analysis for sentiment detection
TL;DR: In this article , the authors analyzed the Amazon dataset using various combinations of voice components and deep learning to identify sentences as "Positive", "Neutral", "Negative", or "Indifferent" and achieved a maximum validation accuracy of 79.83% when using Fast Text as word embedding and the Multi-channel Convolution Neural Network.
23
•Proceedings Article
Designing Truthful Contextual Multi-Armed Bandits based Sponsored Search Auctions
Kumar Abhishek,Shweta Jain,Sujit Gujar +2 more
- 05 May 2020
TL;DR: For sponsored search auctions, this work considers contextual multi-armed bandit problem in the presence of strategic agents as well as the existing solution in this setting as it suffers from very high regret.
4
•Posted Content
Introduction to Concentration Inequalities.
TL;DR: This report aims to exemplify concentration inequalities and provide easy to understand proofs for it, which are helpful in the design and analysis of machine learning algorithms.
Towards fair machine learning using combinatorial methods
TL;DR: A notion of fairness is developed that is based on the notion of discrepancy of set systems, a widely studied topic in the theory of computer science and combinatorics.
3
•Posted Content
Designing Truthful Contextual Multi-Armed Bandits based Sponsored Search Auctions.
TL;DR: In this paper, the authors considered contextual multi-armed bandit problem in the presence of strategic agents and proposed an auction-based solution for sponsored search auctions, where the probability of getting a click depends on the user's profile (context) posting the query.
1