PeopleofSuperintelligence

Researcher

Koby Crammer

Koby Crammer is a machine-learning researcher whose work spans margin-based online learning, domain adaptation, natural language processing and cross-task transfer. At the Hebrew University of Jerusalem, he earned a BSc in 1999 and a computer science PhD in 2004, publishing on multiclass kernel machines with Yoram Singer in 2001, and worked from 2004 to 2009 at the University of Pennsylvania, where he gave an ACL-08 tutorial on online learning for NLP. His papers introduced passive-aggressive online algorithms in 2006, confidence-weighted learning in 2008, Gaussian Margin Machines and the AROW algorithm in 2009, semi-supervised sentiment classification on unlabeled Amazon reviews in 2012, and Target-Aware Weighted Training in 2021. Having joined the Department of Electrical Engineering at the Technion – Israel Institute of Technology in 2009, he is affiliated with the Technion as an artificial intelligence researcher as of October 2026.

Now at
Technion

Timeline

  1. 1999–2004

    Degrees and multiclass kernel machines

    At the Hebrew University of Jerusalem, he receives a BSc in mathematics, physics and computer science in 1999 and a PhD in computer science in 2004, both summa cum laude. In 2001, he and Yoram Singer publish On the Algorithmic Implementation of Multiclass Kernel-based Vector Machines, which gives a direct margin-based method for training multiclass predictors with kernels.

  2. 2006

    Passive-aggressive online learning

    In 2006, he coauthors Online Passive-Aggressive Algorithms with Ofer Dekel, Joseph Keshet, Shai Shalev-Shwartz and Yoram Singer, presenting margin-based online algorithms for binary and multiclass categorization, regression, uniclass prediction and sequence prediction. Also in 2006, he coauthors Analysis of Representations for Domain Adaptation, deriving a generalization bound that illustrates the tradeoffs in designing representations for domain adaptation.

  3. 2008

    NLP research at Penn

    From 2004 to 2009, he serves as a postdoctoral fellow and research associate in the Department of Computer and Information Science at the University of Pennsylvania, holding a PhD from Hebrew University. In 2008, he gives an ACL-08 tutorial on advanced online learning for natural language processing.

  4. 2008

    Confidence-weighted online learning

    In 2008, he coauthors Exact Convex Confidence-Weighted Learning with Mark Dredze and Fernando Pereira. The paper derives a convex constraint for online learning with linear classifiers and analyzes it under a mistake-bound model.

  5. 2009

    Joining Technion and Gaussian Margin Machines

    In 2009, he joins the Department of Electrical Engineering at the Technion – Israel Institute of Technology. In 2009, he also coauthors Gaussian Margin Machines with Mehryar Mohri and Fernando Pereira, maintaining a Gaussian distribution over weight vectors for binary classification, and coauthors Adaptive Regularization of Weight Vectors, introducing the AROW algorithm for online learning under label noise.

  6. 2012

    Sentiment from unlabeled reviews

    In 2012, he coauthors the appendix to More Is Better: Large Scale Partially-supervised Sentiment Classification with Yoav Haimovitch and Shie Mannor. It describes a bootstrapping algorithm for partially labeled data and tests sentiment classification with up to 15 million unlabeled Amazon product reviews.

  7. 2021

    Weighted training across tasks

    In 2021, he coauthors Weighted Training for Cross-Task Learning with Shuxiao Chen, Hangfeng He, Dan Roth and Weijie J. Su. Its Target-Aware Weighted Training method minimizes a representation-based distance between source and target tasks.

  8. As of October 2026

    Research at Technion

    As of October 2026, he is a researcher in artificial intelligence affiliated with the Technion – Israel Institute of Technology.

Career

  1. NowTechnionResearcher
  2. 2008University of PennsylvaniaResearch associate

Education

  1. 2008Hebrew University of JerusalemPhD

From the biography.

  1. Universities keep AI researchers for a median of 6 years, compared with 2 years at frontier labsIncludes Technion. 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 Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes Technion. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.
  3. 41% of people who worked abroad later worked in their country of birth among people with a recorded birth country and dated work historyIncludes Technion. Of 991 people with a recorded birth country who worked outside it, 410 (41%) later held a work position at home, led by the United States at 69%.
  4. The United States employs an estimated 48% of people shaping AIIncludes Technion. 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.

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 Koby Crammer yet. Send us a free-licensed photo.

  1. Learn on Source, Refine on Target: A Model Transfer Learning Framework with Random Forests arxiv.org. Checked 22 October 2026.
  2. On the Algorithmic Implementation of Multiclass Kernel-based Vector Machines jmlr.org. Checked 22 October 2026.
  3. Online Passive-Aggressive Algorithms jmlr.org. Checked 22 October 2026.
  4. Analysis of Representations for Domain Adaptation papers.nips.cc. Checked 22 October 2026.
  5. ACL-08: HLT - Tutorials ling.ohio-state.edu. Checked 22 October 2026.
  6. Exact Convex Confidence-Weighted Learning papers.nips.cc. Checked 22 October 2026.
  7. Gaussian Margin Machines proceedings.mlr.press. Checked 22 October 2026.
  8. Adaptive Regularization of Weight Vectors papers.nips.cc. Checked 22 October 2026.
  9. More Is Better: Large Scale Partially-supervised Sentiment Classification - Appendix arxiv.org. Checked 22 October 2026.
  10. Weighted Training for Cross-Task Learning arxiv.org. Checked 22 October 2026.
  11. Crammer, Koby, Technion - Israel Institute of Technology | TOPSCINET topscinet.com. Checked 22 October 2026.