Ayush Sharma

Shreyansh Singh, Avi Chawla, Ayush Sharma et al.

This paper describes our submission system for the Shallow Track of Surface Realization Shared Task 2018 (SRST'18). The task was to convert genuine UD structures, from which word order information had been removed and the tokens had been lemmatized, into their correct sentential form. We divide the problem statement into two parts, word reinflection and correct word order prediction. For the first sub-problem, we use a Long Short Term Memory based Encoder-Decoder approach. For the second sub-problem, we present a Language Model (LM) based approach. We apply two different sub-approaches in the LM Based approach and the combined result of these two approaches is considered as the final output of the system.

Ayush Sharma, Piyush Bajpai, Sukhdev Singh et al.

Objectives: This paper presents an up-to-date overview of research performed in the Virtual Reality (VR) environment ranging from definitions, its presence in the various fields, and existing market players and their projects in the VR technology. Further an attempt is made to gain an insight on the psychological mechanism underlying experience in using VR device. Methods: Our literature survey is based on the research articles, analysis of the projects of various companies and their findings for different areas of interest. Findings: In our literature survey we observed that the recent advances in virtual reality enabling technologies have led to variety of virtual devices that facilitate people to interact with the digital world. In fact in the past two decades researchers have tried to integrate reality and VR in the form of intuitive computer interface. Improvements: This has led to variety of potential benefits of VR in many applications such as News, Healthcare, Entertainment, Tourism, Military and Defence etc. However despite the extensive research efforts in creating virtual system environments it is yet to become apparent in normal daily life.

Andreas Veit, Michael Wilber, Rajan Vaish et al.

When crowdsourcing systems are used in combination with machine inference systems in the real world, they benefit the most when the machine system is deeply integrated with the crowd workers. However, if researchers wish to integrate the crowd with "off-the-shelf" machine classifiers, this deep integration is not always possible. This work explores two strategies to increase accuracy and decrease cost under this setting. First, we show that reordering tasks presented to the human can create a significant accuracy improvement. Further, we show that greedily choosing parameters to maximize machine accuracy is sub-optimal, and joint optimization of the combined system improves performance.