PeopleofSuperintelligence

Researcher

Lei Feng

Also known as 冯磊

Lei Feng is a machine-learning researcher whose work tackles incomplete and ambiguous labels. He earned a computer science B.E. at Southwest University and a PhD at Nanyang Technological University, then developed partial-label methods that estimate latent labels, use pseudo-labeling, and identify true labels. He coauthored PiCO, which received an ICLR 2022 Honorable Mention, and later coauthored PiCO+, an extension for noisy partial-label learning accepted by IEEE TPAMI in 2023. As of October 2026, he is a full professor at Southeast University and a part-time visiting scientist at RIKEN.

Timeline

  1. 2013–2017

    A computer science degree

    He completes a B.E. in computer science at Southwest University in Chongqing in July 2017.

  2. 2018–2019

    Latent labels and retraining

    In 2018, he and Bo An propose a unified framework that estimates latent label distributions while training a model for partial-label learning. In 2019, they introduce self-guided retraining, which trains a model while performing pseudo-labeling on partially labeled examples.

  3. 2020

    Identifying true labels

    In 2020, he coauthors “Progressive Identification of True Labels for Partial-Label Learning,” a method that updates the model while identifying true labels from candidate sets.

  4. 2021

    A PhD and thesis award

    He receives his PhD in computer science from Nanyang Technological University in February 2021 after defending his thesis in January 2021. He is runner-up for the NTU SCSE Outstanding PhD Thesis Award.

  5. 2022

    The PiCO framework

    In 2022, he coauthors PiCO with six collaborators. The paper receives an ICLR 2022 Honorable Mention, and its framework combines contrastive learning with class-prototype label disambiguation.

  6. 2023

    Extending PiCO to noisy labels

    In 2023, PiCO+ is accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence, with Feng among its seven authors. It extends PiCO with distance-based clean-sample selection and semi-supervised contrastive learning for noisy partial labels.

  7. As of October 2026

    Professor at Southeast University

    As of October 2026, he is a full professor at PALM Lab in the School of Computer Science and Engineering at Southeast University, and a part-time visiting scientist at RIKEN’s Center for Advanced Intelligence Project.

Career

  1. NowSoutheast UniversityFull Professor
  2. NowRIKENPart-time Visiting Scientist

Education

  1. 2021Nanyang Technological UniversityPhD
  2. 2017Southwest UniversityB.E.

From the biography.

  1. Universities keep AI researchers for a median of 6 years, compared with 2 years at frontier labsIncludes RIKEN. Across 15,811 dated work stays (6,529 people), median role tenure runs from 6 years at universities to 2 years at frontier labs. We compare all five organization types, starting decades from the 1970s to the 2020s, and the organizations that keep people longest.
  2. The United States employs an estimated 48% of people shaping AIIncludes RIKEN. We estimate where the people shaping AI work today, how universities, companies and research institutes split across countries, and how country shares have changed since the 1960s.
  3. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes RIKEN. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  4. Google has the highest combined h-index among AI organizationsIncludes RIKEN. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Lei Feng yet. Send us a free-licensed photo.

  1. Lei Feng lfeng1995.github.io. Checked 21 October 2026.
  2. Leveraging Latent Label Distributions for Partial Label Learning | IJCAI ijcai.org. Checked 21 October 2026.
  3. Partial Label Learning with Self-Guided Retraining | Proceedings of the AAAI Conference on Artificial Intelligence ojs.aaai.org. Checked 21 October 2026.
  4. Progressive Identification of True Labels for Partial-Label Learning proceedings.mlr.press. Checked 21 October 2026.
  5. ICLR 2022 Awards iclr.cc. Checked 21 October 2026.
  6. PiCO: Contrastive Label Disambiguation for Partial Label Learning hbzju.github.io. Checked 21 October 2026.
  7. Lei Feng lfeng1995.github.io. Checked 21 October 2026.
  8. PiCO+: Contrastive Label Disambiguation for Robust Partial Label Learning arxiv.org. Checked 21 October 2026.