Steve Hanneke

Computer Science · Purdue University West Lafayette

Mid career · publishing since 2004Rising activity

Machine Learning and Algorithms Advanced Bandit Algorithms Research Domain Adaptation and Few-Shot Learning Algorithms and Data Compression Machine Learning and Data Classification

Publications

162

Citations

2,415

Est. group size

—

Recurring co-author estimate

Active years

Publishing since 2004

Research summary

AI-generated

Steve Hanneke works in the theoretical foundations of machine learning, focusing on questions like how much data is needed to learn accurately (sample complexity), how learning algorithms behave under uncertainty or adversarial conditions, and the mathematical limits of what can be learned in settings such as multitask learning, online learning, and reasoning with AI models. This work is mostly mathematical and aims to establish provable guarantees and limitations for learning algorithms rather than building specific applied systems. Students interested in rigorous, proof-based analysis of machine learning theory would find this area relevant.

Statistical learning theory and sample complexityOnline and bandit learning algorithmsAgnostic and adversarial learning modelsMultitask and domain adaptation theoryFoundations of classification and reasoning in AI

Publication output has grown over the past decade, rising from single digits per year in 2017-2020 to over 20 papers annually in 2024 and 2026, indicating an increasing and currently high level of research activity.

Generated by claude-sonnet-5 from public bibliographic data · Jul 20, 2026

Publication cadence

Publications per year over the last 10 years — averaging 16.8/year recently

Recent publications

On the Learning Curves of Revenue Maximization
arXiv (Cornell University) · 2026
Sample Complexity of Agnostic Multiclass Classification: Natarajan Dimension Strikes Back
2026
When More Data Doesn't Help: Limits of Adaptation in Multitask Learning
Open MIND · 2026
When More Data Doesn't Help: Limits of Adaptation in Multitask Learning
arXiv (Cornell University) · 2026
A Theory of Universal Agnostic Learning
Open MIND · 2026
A Theory of Universal Agnostic Learning
arXiv (Cornell University) · 2026
Sample Complexity of Autoregressive Reasoning: Chain-of-Thought vs. End-to-End
arXiv (Cornell University) · 2026
An Optimal Sauer Lemma Over $k$-ary Alphabets
arXiv (Cornell University) · 2026
Sample Complexity of Autoregressive Reasoning: Chain-of-Thought vs. End-to-End
arXiv (Cornell University) · 2026
An Optimal Sauer Lemma Over $k$-ary Alphabets
arXiv (Cornell University) · 2026
On the Learning Curves of Revenue Maximization
arXiv (Cornell University) · 2026
Realizable Bayes-Consistency for General Metric Losses
arXiv (Cornell University) · 2026
Realizable Bayes-Consistency for General Metric Losses
arXiv (Cornell University) · 2026
Regret-Oracle Complexity Tradeoffs in Agnostic Online Learning
arXiv (Cornell University) · 2026
A Complete Characterization of Learnability for Adversarial Noisy Bandits
arXiv (Cornell University) · 2026

Publishes in

arXiv (Cornell University)×80
Theoretical Computer Science×5
Algorithmic Learning Theory×3
Conference on Learning Theory×3
The Annals of Statistics×2

Similar researchers

By research-topic overlap

Changlong Wu
Computer Science · Purdue University West Lafayette
Paul Valiant
Computer Science · Purdue University West Lafayette
Roni Khardon
Computer Science · Indiana University
Tianyu Wang
Decision Sciences · The Ohio State University
Net Zhang
Decision Sciences · The Ohio State University

This profile was generated automatically from public scholarly data (OpenAlex). Group size and activity levels are estimates derived from co-authorship patterns.

Last updated Jul 20, 2026.

Claim or correct this profile