Mehryar Mohri

Mehryar Mohri is known for applying weighted automata and finite-state transducers to speech recognition and natural language processing. His algorithms for constructing and optimizing these mathematical structures became foundational components in practical speech and language systems used at companies like Google.

Mehryar Mohri is a professor at New York University's Courant Institute of Mathematical Sciences and is also affiliated with Google Research. His dual academic and industry roles have helped bridge theoretical computer science with large-scale machine learning applications.

Mehryar Mohri's research on weighted automata and finite-state transducers provides efficient algorithms for representing and manipulating sequences that carry associated costs or probabilities. These structures are used in speech recognition, where both acoustic models and language models can be represented as weighted automata and combined efficiently.

Mehryar Mohri co-authored 'Foundations of Machine Learning,' a widely used graduate-level textbook that provides a rigorous mathematical treatment of learning theory, generalization bounds, and core algorithms. The book is used in computer science programs at numerous universities.