Corinna Cortes

Vice President at Google Research & Co-Inventor of Support Vector Machines

Lifespan 1961–

Technical Depth advanced

Key Breakthrough Soft-Margin Support Vector Machine (SVM) Algorithm (1995)

Focus Areas

machine learning algorithms theoretical computer science

Topic Keywords

#machine-learning #svm #convex-optimization #algorithms #google-research

Source: Historical Biographical Archive / Wikimedia Commons

Corinna Cortes

Vice President at Google Research & Co-Inventor of Support Vector Machines

Biographical Overview

Vice President at Google Research and former Head of Google Research NY. At AT&T Bell Labs in 1995, Cortes co-invented the soft-margin Support Vector Machine (SVM) with Vladimir Vapnik, creating one of the most celebrated and foundational algorithms in machine learning history.

"The beauty of SVMs is that they balance mathematical elegance with practical robustness on real-world noisy data."
— Corinna Cortes

Lifespan 1961–

Technical Depth advanced

Key Breakthrough Soft-Margin Support Vector Machine (SVM) Algorithm (1995)

Focus Areas

machine learning algorithms theoretical computer science

Topic Keywords

#machine-learning #svm #convex-optimization #algorithms #google-research

Source: Historical Biographical Archive / Wikimedia Commons

💡

Historical Context & Impact

In short

Cortes and Vapnik's 1995 paper "Support-Vector Networks" in Machine Learning is among the top five most cited papers in all of computer science. For fifteen years until the rise of modern deep learning, SVMs were the gold standard algorithm in machine learning.

Key Technical Breakthroughs & Inventions

Soft-Margin Support Vector Machine Formulation (1995) Co-invented soft-margin SVMs by introducing slack variables into convex quadratic optimization, allowing classifiers to handle noisy, non-linearly separable data.

Kernel Methods and Generalization Bounds Established theoretical VC-dimension and Rademacher complexity bounds for kernel classifiers, regression trees, and ensemble boosting algorithms.

Leadership of Google Research NY Scaled Google's New York research lab into an international center for algorithmic auction design, natural language processing, and speech recognition.

Algorithmic Foundations of Handwriting Recognition Designed the high-throughput neural and kernel recognition systems used to process millions of US postal envelopes and bank checks automatically.