Arun Nampally

Arun Nampally

Staff AI Scientist

Invitae

Biography

I am a computer scientist with a background in logic, probabilistic inference/learning and artificial intelligence. Currently at Invitae, I apply probabilistic machine learning models for challenging problems in computational biology. Some of the projects that I have worked on include developing a probabilistic model for variant interpretation and developing an algorithm to leverage our sequence database together with population genetics to impute missing data for targeted panels.

Beyond the application of ML to biology/genetics, I am interested ML formalisms that leverage structured representations of domain knowledge in the form of graphs and logical rules/theories – statistical relational learning, probabilistic (logic) programming etc. I have expertise in related tools such as ProbLog and Pyro.

Interests
  • Artificial Intelligence
  • Machine Learning
  • Probabilistic Programming
  • Probabilistic Logic Programming
  • Statistical Relational Learning
Education
  • PhD in Computer Science, 2018

    Stony Brook University

Publications

(2021). Graphical Models For Rare Sequence Variant Interpretation. In Trustworthy AI for Healthcare (AAAI2021 Workshop).

PDF Poster Workshop

(2018). Constraint-Based Inference in Probabilistic Logic Programs. TPLP.

PDF Cite DOI

(2016). Inference in Probabilistic Logic Programs using Lifted Explanations. In ICLP 2016.

PDF Cite DOI

(2015). Constraint-Based Inference in Probabilistic Logic Programs. In PLP 2015.

PDF Cite workshop

(2014). Adaptive MCMC-based inference in probabilistic logic programs. In ICLP 2014.

PDF Cite DOI