Gautam Kamath

[ALT 2025] Sample Compression Scheme Reductions

[ALT 2025] Do PAC-Learners Learn the Marginal Distribution?

[ALT 2025] Plenary #1: AI safety via Inference-time compute (Boaz Barak)

[ALT 2025] Efficient Optimal PAC Learning

[ALT 2025] Is Transductive Learning Equivalent to PAC Learning?

[ALT 2025] Cost-Free Fairness in Online Correlation Clustering

[ALT 2025] Boosting, Voting Classifiers and Randomized Sample Compression Schemes

[ALT 2025] Minimax-optimal and Locally-adaptive Online Nonparametric Regression

[ALT 2025] Plenary #2: Linear Operators Learning for Dynamical Systems (Massimiliano Pontil)

[ALT 2025] Clustering with bandit feedback: breaking down the computation/information gap

[ALT 2025] Sharp bounds on aggregate expert error

[ALT 2025] Understanding Aggregations of Proper Learners in Multiclass Classification

[ALT 2025] A Complete Characterization of Learnability for Stochastic Noisy Bandits

[ALT 2025] Quantile Multi-Armed Bandits with 1-bit Feedback

[ALT 2025] Logarithmic Regret for Unconstrained Submodular Maximization Stochastic Bandit

[ALT 2025] Computationally efficient reductions between some statistical models

[ALT 2025] Agnostic Private Density Estimation for GMMs via List Global Stability

[ALT 2025] Info-Theoretic Guarantees for Recovering Low-Rank Tensors from Symm. Rank-1 Measurements

[ALT 2025] On the Hardness of Learning One Hidden Layer Neural Networks

[ALT 2025] Hi-acc sampling from constrained spaces w/ Metropolis-adjusted Precond Langevin Alg

[ALT 2025] On Generalization Bounds for Neural Networks with Low Rank Layers

[ALT 2025] When and why randomised exploration works (in linear bandits)

[ALT 2025] Fast Convergence of Φ-Divergence Along Unadjusted Langevin Algorithm & Proximal Sampler

[ALT 2025] Differentially Private Multi-Sampling from Distributions

[ALT 2025] Optimal Rates for O(1)-Smooth DP-SCO with a Single Epoch and Large Batches

[ALT 2025] Center-Based Approximation of a Drifting Distribution

[ALT 2025] Noisy Computing of the Threshold Function

[ALT 2025] A Unified Theory of Supervised Online Learnability

[ALT 2025] Nearly-tight Approximation Guarantees for the Improving Multi-Armed Bandits Problem

[ALT 2025] Non-stochastic Bandits With Evolving Observations

[ALT 2025] Full Swap Regret and Discretized Calibration

[ALT 2025] Online Learning of Quantum States with Logarithmic Loss via VB-FTRL

[ALT 2025] Efficient PAC Learning of Halfspaces with Constant Malicious Noise Rate

[ALT 2025]: Plenary #3: Estimation to Prediction: What Assumptions Do We Need? (Nikita Zhivotovskiy)

[ALT 2025] For Universal Multiclass Online Learning, Bandit Feedback & Full Supervision are Equiv.

[ALT 2025] Generalisation under gradient descent via deterministic PAC-Bayes

[ALT 2025] Interview with Nicolò Cesa-Bianchi

[ALT 2025] Dimension Strikes Back w/ Gradients: Generalization of Gradient Methods in Stoch Conv Opt

[ALT 2025] Generalization bounds for mixing processes via delayed online-to-PAC conversions

[ALT 2025] A Model for Combinatorial Dictionary Learning and Inference

[ALT 2025] A PAC-Bayesian Link Between Generalisation and Flat Minima

[ALT 2025] How rotation invariant algorithms are fooled by noise on sparse targets

[ALT 2025] Enhanced $H$-Consistency Bounds

[ALT 2025] An Online Feasible Point Method for Benign Generalized Nash Equilibrium Problems

[ALT 2025] Strategyproof Learning with Advice

[ALT 2025] Plugin Approach for Average-Reward & Discounted MDPs: Optimal Sample Complexity Analysis

[ALT 2025] Optimal and learned algorithms for the online list update problem with Zipfian accesses

[ALT 2025] Self-Directed Node Classification on Graphs

[ALT 2025] Proper Learnability and the Role of Unlabeled Data

[ALT 2025] Reliable Active Apprenticeship Learning

[ALT 2025] Effective Littlestone dimension

[ALT 2025] Error dynamics of mini-batch gradient descent w/ random reshuffling for least squares

[ALT 2025] A Characterization of List Regression

[ALT 2025] Refining the Sample Complexity of Comparative Learning

[ALT 2025] Plenary #4: RL beyond expectations: Planning for utility functions (Claire Vernade)

CS 480/680 - Lecture 20 - Ethics

CS 480/680 - Lecture 19 - Attention

CS 480/680 - Lecture 18 - Privacy

CS 480/680 - Lecture 17 - Robustness

CS 480/680 - Lecture 16 - Generative Adversarial Networks

Lecture 8A: Private Multiplicative Weights - Linear Queries

Lecture 7A: The Exponential Mechanism

Lecture 6A: Advanced Composition - Reduction to Binary-ish Case

Lecture 6B: Advanced Composition - Composition Proof

Lecture 4B: Intro to Differential Privacy, Part 2 - Properties of Differential Privacy

Lecture 3A: Randomized Response

Lecture 2A: Reconstruction Attacks - How to Attack a Census

Lecture 2C: Reconstruction Attacks - Cohen-Nissim attack Diffix!

Lecture 1A: Some Attempts at Data Privacy - NYC Taxis and Netflix

Lecture 1B: Some Attempts at Data Privacy - Neural Networks, Medical Studies, k-Anonymity