Quoc V. Le

Lê Viết Quốc (born 1982),^[1] or in romanized form Quoc Viet Le, is a Vietnamese-American computer scientist and artificial intelligence researcher. He is a Google Fellow at Google DeepMind and a founding member of the Google Brain project.

Le is best known for his pioneering work in deep learning, particularly in large-scale unsupervised learning, Sequence-to-Sequence (seq2seq^[2]) models, and AutoML (neural architecture search). His research laid the foundation for modern machine translation systems, such as Google Translate, and advanced the field of Large Language Models (LLMs).^[3][4][5][6]

Le was born in Hương Thủy in the Thừa Thiên Huế province of Vietnam.^[4] He attended Quốc Học Huế High School^[7] before moving to Australia in 2004 to pursue a Bachelor’s degree at the Australian National University. During his undergraduate studies, he worked with Alex Smola on Kernel method in machine learning.^[8] In 2007, Le moved to the United States to pursue graduate studies in computer science at Stanford University, where his PhD advisor was Andrew Ng.

In 2011, Le became a founding member of Google Brain along with his then advisor Andrew Ng, Google Fellow Jeff Dean, and researcher Greg Corrado.^[4] He led Google Brain’s first major breakthrough: a deep learning algorithm trained on 16,000 CPU cores, which learned to recognize cats by watching YouTube videos—without being explicitly taught the concept of a "cat."^[9][10]

In 2014, Le co-proposed two influential models in machine learning. Together with Ilya Sutskever, Oriol Vinyals, he introduced the seq2seq model for machine translation, a foundational technique in natural language processing. In the same year, in collaboration with Tomáš Mikolov, Le developed the doc2vec^[11] model for representation learning of documents. Le was also a key contributor of Google Neural Machine Translation system.^[12]

In 2017, Le initiated and led the AutoML project at Google Brain, pioneering the use of neural architecture search.^[13] This project significantly advanced automated machine learning. This work led to EfficientNet, a family of image recognition models that achieved state-of-the-art accuracy while being significantly smaller and faster than previous models.

In 2020, Le contributed to the development of Meena, later renamed LaMDA, a conversational large language model based on the seq2seq architecture.^[14] In 2022, Le and coauthors published chain-of-thought prompting, a method that enhances the reasoning capabilities of large language models.^[15]

In 2024, Le contributed to the development of AlphaGeometry, an AI system that solves complex geometry problems at a level approaching a human International Mathematical Olympiad (IMO) gold-medalist. The system, published in Nature, demonstrated the ability to solve 25 out of 30 Olympiad geometry problems, significantly outperforming previous state-of-the-art automated theorem provers.^[16]

Le was named MIT Technology Review's innovators under 35 in 2014.^[17] He has been interviewed by and his research has been reported in major media outlets including Wired,^[5] the New York Times,^[18] the Atlantic,^[19] and the MIT Technology Review.^[20] His work on the 2012 Google Brain project, which pioneered large-scale unsupervised learning, received an ICML Test of Time Honorable Mention Award in 2022.^[21] Le was named an Alumni Laureate of the Australian National University School of Computing in 2022.^[22] In 2024, Le received the NeurIPS Test of Time Award for the seminal paper "Sequence to Sequence Learning with Neural Networks" (co-authored with Ilya Sutskever and Oriol Vinyals). The award committee described the work as a "cornerstone" of modern AI, noting that it established the encoder-decoder architecture and laid the necessary foundation for the large language models (LLMs) and foundation models that followed.^[23]

• Oriol Vinyals
• Ilya Sutskever
• Jeff Dean
• Alex Smola