Adhi Saravanan

Adhi Saravanan

Computer Engineering Student

University of Cambridge

About Me

Currently, I am an MEng student at the University of Cambridge and I am advised by Matthew Wicker and Adrian Weller. My thesis research is focussed on formal methods for the certification of properties of Bayesian neural networks. Previously, I spent a summer under Professor Anima Anandkumar and Rafal Kocielnik at the Anima AI+Science Lab, at Caltech. Here, I worked on applied ML problems - under the theme of automatic bias discovery in LLMs and text-to-image models.

I am grateful for on-going collaboration with amazing researchers at the Anima AI + Science Lab and the DeepWok Lab.

Outside of computer science, I enjoy playing cricket. I have played for representative teams at the University of Cambridge and even played three matches for Caltech, in the Orange County Cricket Association!

Interests
  • Trustworthy Machine Learning (Robustness, Uncertainty and Interpretability)
  • Theory (Learning, Generalization)
Education
  • BA, MEng in Information and Computer Engineering, 2020 - 2024

    University of Cambridge

Education

 
 
 
 
 
University of Cambridge (Selwyn College)
Master of Engineering (BA, MEng)
October 2020 – June 2024 Cambridge, Cambridgeshire
  • Master’s Classes: Probabilistic Machine Learning, Deep Learning and Structured Data, Probabilisic Machine Learning, Practical Optimisation, Computational Neuroscience, Materials and Molecules: Modelling, Simulation and Machine Learning
  • Year 3: Class I (Top 3%)
  • Year 2: Class I (Top 10%)
  • Year 1: Class I (No rank due to COVID)
  • Awards: Harrison Award (Top-of-the-class in Selwyn Engineering - 2021, 2023), Christopher Johnson Award (academic and extra-curricular contributions to life at Selwyn College, 2 awardees out of ~ 500 per year - 2023), Selwyn College Scholarship and Tripos Prize (Class I result - 2021, 2022, 2023)
 
 
 
 
 
Greenhead College
A-Levels
October 2018 – June 2020 Huddersfield, West Yorkshire
  • 4A*s: Mathematics, Further Mathematics, Chemistry, Physics
 
 
 
 
 
Heckmondwike Grammar School
GCSEs
October 2013 – June 2018 Heckmondwike, West Yorkshire
  • 8 Grade 9s, 2A*s, 1 Grade 8 and an A in Additional Mathematics FSMQ
  • 3 gold awards in the UKMT Math Challenge - 2 x best in school, 1 x Qualification for the Math Olympiad (Merit)
  • 6 subject awards for top-4 (/180) performance in graduating Year 11 class including English (99.4%), History and Biology

Experience

 
 
 
 
 
Anima AI + Science Lab (Caltech)
Machine Learning Research Intern
June 2023 – September 2023 Pasadena, California
 
 
 
 
 
Bolt6.ai
Software Engineering Intern
June 2022 – September 2022 London, United Kingdom
 
 
 
 
 
Optalysys
Machine Learning Applications Intern
June 2021 – September 2021 Leeds, United Kingdom

Publications

(2023). Empowering Domain Experts to Detect Social Bias in Generative AI with User-Friendly Interfaces. XAI in Action: Past, Present, and Future Applications.

Cite URL

(2023). Exploring Social Bias in Downstream Applications of Text-to-Image Foundation Models. I Can’t Believe It’s Not Better Workshop: Failure Modes in the Age of Foundation Models.

Cite URL