Faculty

Institution

Unit

Venue

Nathan Srebro

Professor Part-Time

Toyota Technological Institute at Chicago

Affiliations

Toyota Technological Institute at Chicago

Nathan Srebro is currently serving as a Toyota Technological Institute At Chicago Professor Part-Time at Toyota Technological Institute At Chicago.

Honors

Create free account to unlock more features

Experiences

Create free account to unlock more features

Education

Create free account to unlock more features

Nathan Srebro

Toyota Technological Institute at Chicago

Professor Part-Time

You can click a keyword to filter the displayed items

machine learning computer science applied mathematics machine learning research data science mathematical optimization statistics supervised learning deep learning stochastic optimization computational intelligence statistical learning theory artificial intelligence computational optimization computational statistics computational learning theory approximation theory learning theory stochastic process natural language processing convex optimization adversarial machine learning statistical inference computer vision convex analysis distributed learning kernel methods implicit bias distributed computing support vector machines mathematical programming numerical algorithms semi-supervised learning regularization (mathematics) stochastic gradient descent neural networks combinatorial optimization online algorithms neural scaling law stochastic convex optimization implicit regularization uniform convergence supervised classification separable data automated reasoning function spaces distributed systems scientific computing social science research machine learning models probabilistic graph theory constrained optimization fair data principle algorithmic fairness infinite dimensional analysis infinite dimensional problem distributed processing matrix factorization sparse representations cost-sensitive machine learning randomized kaczmarz algorithm empirical risk minimization empirical minimization path-normalized optimization similarity functions fast rates gradient descent optimistic rates statistical methodology evolutionary computation collaborative filtering graphical models inductive inference optimization problems algorithmic bias parametric programming

Notable Scholarly Works

Notable Federal Grants

Collaborators

Overfitting and Generalizing With (Pac) Bayesian Prediction in Noisy Binary Classification. arXiv (Cornell University) (2026)

bayesian inferencebayesian predictionnoisy binary classificationclassification methodnoisy datastatistical inferencebayesian analysisstatisticsinferencemachine learning researchdata sciencecomputational intelligencecomputational statisticsmachine learningmodel selectioncomputer science

A Theory of Learning With Autoregressive Chain of Thought. ArXiv.org (2025)

learning theoryapproximate reasoningreinforcement learningcognitive sciencedata-driven learningmachine learningcognition

From Continual Learning to Sgd and Back: Better Rates for Continual Linear Models. ArXiv.org (2025)

stochastic optimizationmathematical optimizationreinforcement learningsequential learningmachine learning researchcomputational intelligenceapplied mathematicsmachine learning

Learning Single-Index Models via Harmonic Decomposition. ArXiv.org (2025)

multilinear subspace learninglow-rank approximationmachine learning researchharmonic decompositionapplied mathematicscomputational statisticsmachine learningmodel selection

Learning to Answer From Correct Demonstrations. arXiv (Cornell University) (2025)

correct demonstrationsimitative learningreinforcement learningrobot learninginferenceautonomous learningaction model learninglanguagecomputer science

On the Hardness of Learning Regular Expressions. ArXiv.org (2025)

computational complexitycombinatorial pattern matchingknowledge discoverycomputational complexity theorycomputer science

Quantifying Overfitting Along the Regularization Path for Two-Part-Code MDL in Supervised Classification. ArXiv.org (2025)

machine learningsupervised classificationtwo-part-code mdlclassification methodcost-sensitive machine learningregularization pathmachine learning researchpattern recognitiondata sciencesupervised learningcomputer scienceautomated machine learning

Temperature is All You Need for Generalization in Langevin Dynamics and Other Markov Processes. ArXiv.org (2025)

physicsapplied probabilitygibbs measurestatistical physicsstochastic processapplied mathematicscomputational physicsprobability theory

Tight Bounds on the Binomial Cdf, and the Minimum of I.I.D Binomials, in Terms of Kl-Divergence. ArXiv.org (2025)

lower boundsmathematical statisticstight boundsbinomial cdfstatisticsstochastic processasymptotic analysisapplied mathematicsi.i.d binomialsprobability theory

Weak-To-Strong Generalization Even in Random Feature Networks, Provably. ArXiv.org (2025)

neural network (machine learning)machine learningmathematical optimizationmachine learning researchknowledge discoverycomputer scienceadversarial machine learning

1
2
3
4
5
17
10 / page

Nathan Srebro

Professor Part-Time

Toyota Technological Institute at Chicago

Affiliations

Toyota Technological Institute at Chicago

Nathan Srebro is currently serving as a Toyota Technological Institute At Chicago Professor Part-Time at Toyota Technological Institute At Chicago.

Honors

Create free account to unlock more features

Experiences

Create free account to unlock more features

Education

Create free account to unlock more features

Nathan Srebro

Toyota Technological Institute at Chicago

Professor Part-Time

You can click a keyword to filter the displayed items

Notable Scholarly Works

Notable Federal Grants

Collaborators

Overfitting and Generalizing With (Pac) Bayesian Prediction in Noisy Binary Classification. arXiv (Cornell University) (2026)

A Theory of Learning With Autoregressive Chain of Thought. ArXiv.org (2025)

learning theoryapproximate reasoningreinforcement learningcognitive sciencedata-driven learningmachine learningcognition

From Continual Learning to Sgd and Back: Better Rates for Continual Linear Models. ArXiv.org (2025)

stochastic optimizationmathematical optimizationreinforcement learningsequential learningmachine learning researchcomputational intelligenceapplied mathematicsmachine learning

Learning Single-Index Models via Harmonic Decomposition. ArXiv.org (2025)

multilinear subspace learninglow-rank approximationmachine learning researchharmonic decompositionapplied mathematicscomputational statisticsmachine learningmodel selection

Learning to Answer From Correct Demonstrations. arXiv (Cornell University) (2025)

correct demonstrationsimitative learningreinforcement learningrobot learninginferenceautonomous learningaction model learninglanguagecomputer science

On the Hardness of Learning Regular Expressions. ArXiv.org (2025)

computational complexitycombinatorial pattern matchingknowledge discoverycomputational complexity theorycomputer science

Quantifying Overfitting Along the Regularization Path for Two-Part-Code MDL in Supervised Classification. ArXiv.org (2025)

Temperature is All You Need for Generalization in Langevin Dynamics and Other Markov Processes. ArXiv.org (2025)

physicsapplied probabilitygibbs measurestatistical physicsstochastic processapplied mathematicscomputational physicsprobability theory

Tight Bounds on the Binomial Cdf, and the Minimum of I.I.D Binomials, in Terms of Kl-Divergence. ArXiv.org (2025)

lower boundsmathematical statisticstight boundsbinomial cdfstatisticsstochastic processasymptotic analysisapplied mathematicsi.i.d binomialsprobability theory

Weak-To-Strong Generalization Even in Random Feature Networks, Provably. ArXiv.org (2025)

neural network (machine learning)machine learningmathematical optimizationmachine learning researchknowledge discoverycomputer scienceadversarial machine learning

1
2
3
4
5
17
10 / page