Vlad Niculae
Niculae coauthored the scikit-learn API-design paper. The team described software interfaces and design practices for reusable machine-learning tools. He also worked on core development of scikit-learn. Niculae earned a Cornell PhD in 2018 with Claire Cardie, followed by postdoctoral work in Lisbon. His research focuses on sparsity, structure and geometry in language learning. Niculae coauthored Discrete Latent Structure in Neural Networks. The work reviews methods for learning hidden discrete structures with continuous relaxation, surrogate gradients and probabilistic estimation.
- Now at
- University of Amsterdam
- Works
- 70
- Citations
- 3,280
- h-index
- 16
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2013
Machine-learning software
Niculae coauthored the scikit-learn API-design paper. The team described software interfaces and design practices for reusable machine-learning tools. He also worked on core development of scikit-learn.
-
2018
Cornell doctoral research
Niculae earned a Cornell PhD in 2018 with Claire Cardie, followed by postdoctoral work in Lisbon. His research focuses on sparsity, structure and geometry in language learning.
-
2023
Discrete structure in neural models
Niculae coauthored Discrete Latent Structure in Neural Networks. The work reviews methods for learning hidden discrete structures with continuous relaxation, surrogate gradients and probabilistic estimation.
Sources and credits
Photo: hybrid-intelligence-centre.nl; Vlad Niculae – The Hybrid Intelligence Centre; creator not stated. License: Not stated by source. Source: hybrid-intelligence-centre.nl. Cropped to a square and resized. Ask us to remove or replace this photo.