Vishwas MruthyunjayaMegagon Labs · Research
Vishwas Mruthyunjaya
Master of Robotics Technology
About
9
Publications
1,054
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
23
Citations
Introduction
Applied Scientist in conversational AI and social robotics. Research focus in computational linguistics, spoken dialogue systems, and human-robot interaction (HRI). 5+ years of hands-on experience designing end-to-end human-interactive intelligent systems for conversational AI platforms and physical robots using dynamic predictive models, language models, and deep learning models on real-life data. Thoroughly enjoys researching and developing ethically capable intelligent systems.
Publications
Publications (9)
Symbolic knowledge graphs (KGs) play a pivotal role in knowledge-centric applications such as search, question answering and recommendation. As contemporary language models (LMs) trained on extensive textual data have gained prominence, researchers have extensively explored whether the parametric knowledge within these models can match up to that p...
Science and technology drive innovation, create economic opportunity, and are critical to national security. With increased competition for a skilled STEM workforce, high barriers to participation in STEM, the missing millions (Gershenfeld et al., 2021), and the longstanding underrepresentation of minoritized US communities, collective action is ur...
This paper provides a system design framework for a human-robot interaction system. The design introduces a human-augmented robotic intelligence embedded in a human-robot interaction system. The motivations behind the system design are spoken dialogue systems, Wizard-of-OZ framework, and existing HRI designs for socially intelligent robots. In this...
The aim of this paper is to mold and materialize the future of learning. The paper introduces a Modular Cognitive Educator System (MCES), which aims to help people learn cognitive and ethical capabilities to face one of the indirect impacts of the robot revolution, namely, its impact on the educational system. MCES discusses the importance of agile...
Social robots deployed in public spaces present a challenging task for ASR because of a variety of factors, including noise SNR of 20 to 5 dB. Existing ASR models perform well for higher SNRs in this range, but degrade considerably with more noise. This work explores methods for providing improved ASR performance in such conditions. We use the AiSh...
In this paper we determine how multi-layer ensembling improves performance on multilingual intent classification. We develop a novel multi-layer ensembling approach that ensembles both different model initializations and different model architectures. We also introduce a new banking domain dataset and compare results against the standard ATIS datas...
The main goal of this research is to classify human facial expressions in a human-robot interaction. The project used an existing robotic system, Socibot – a social robot developed by Engineered Arts, to implement the facial expression classifier. The target is to pick the visual cues and recognise the mental or emotional state of the user. This cl...
