Vyas Raina


Hi, I am Vyas and welcome to my page! I am currently a third year PhD student in the Machine Intelligence Lab at the University of Cambridge, United Kingdom. In a world where increasingly powerful artificial intelligence (AI) models are performing some of the most complex tasks with super-human ability the future is fast and exciting. However, if I may quote an age-old adage (famously borrowed later by Marvel), with great power comes great responsibility; i.e, the better our AI models, the greater the need for safe AI. Hence, with a focus mainly in the domain of natural language, my research seeks to use adversarial attacks, defence and robustness of deep learning models as a primary lens of analysis to further the field of safe AI for models deployed in high stakes environments. Do explore my publications and other related projects in the Research section.

Beyond research, I do maintain a foot in a range of industrial projects - check the Work tab. My experiences have allowed me to apply machine learning methods to a range of sectors: Natural Language Processing (NLP) in finance; NLP forecasting in telecommunications; uncertainty methods for tabular weather data; testing for a AI robotics' company and finally I have co-founded a digitial healthcare start-up.

Quick links: CV, Google scholar, LinkedIn

News


  • UK Speech Conference Sep 2022
    Poster presentation of paper, Gender Bias and Universal Substitution Adversarial Attacks on Grammatical Error Correction Systems for Automated Assessment, at UK Speech, Edinburgh.
  • Start-up: MedPredict Aug 2022
    Incorporated (with Companies House) digital healthcare startup, MedPredict
  • NAACL Conference Jul 2022
    Poster presentation of paper, Residue-Based Natural Language Adversarial Attack Detection, at NAACL, Seattle.
  • SHIFTS 1.0 Challenge, NeurIPS Dec 2021
    Organized and ran Shifts 1.0 Challenge, at NeurIPS 2021 (virtual)
  • Schonfeld Oct 2021 - Now
    Began part-time work as NLP research scientist at hedgefund, Schonfeld
  • Interspeech Conference Oct 2020
    Virtual Oral Presentation of my paper Universal Adversarial Attacks on Spoken Language Assessment Systems, at INTERSPEECH Shanghai.
  • PhD Oct 2020 - Now
    Began my PhD in the Machine Intelligence Lab, University of Cambridge, supervised by Prof. Mark Gales.