The proceedings of COLT 2017 are available in the Proceedings of Machine Learning Research (PMLR)
Best Paper. A Hitting Time Analysis of Stochastic Gradient Langevin Dynamics
Best Student Paper. Online Learning Without Prior Information
Ten Steps of EM Suffice for Mixtures of Two Gaussians
Noisy Population Recovery from Unknown Noise
Empirical Risk Minimization for Stochastic Convex Optimization: $O(1/n)$- and $O(1/n^2)$-type of Risk Bounds
Lower Bounds on Regret for Noisy Gaussian Process Bandit Optimization
Efficient Co-Training of Linear Separators under Weak Dependence
Mixing Implies Lower Bounds for Space Bounded Learning
The Price of Selection in Differential Privacy
Learning Disjunctions of Predicates
Submodular Optimization under Noise
Two-Sample Tests for Large Random Graphs using Network Statistics
A second-order look at stability and generalization
The Sample Complexity of Optimizing a Convex Function
An Improved Parametrization and Analysis of the EXP3++ Algorithm for Stochastic and Adversarial Bandits
Ignoring Is a Bliss: Learning with Large Noise Through Reweighting-Minimization
Corralling a Band of Bandit Algorithms
Surprising properties of dropout in deep networks
Optimal learning via local entropies and sample compression
Learning Multivariate Log-concave Distributions
The Simulator: Understanding Adaptive Sampling in the Moderate-Confidence Regime
Quadratic Upper Bound for Recursive Teaching Dimension of Finite VC Classes
Fast Rates for Empirical Risk Minimization of Strict Saddle Problems
Sampling from a log-concave distribution with compact support with proximal Langevin Monte Carlo
Fundamental limits of symmetric low-rank matrix estimation
Homotopy Analysis for Tensor PCA
The Hidden Hubs Problem
Depth Separation for Neural Networks
Bandits with Movement Costs and Adaptive Pricing
Nearly-tight VC-dimension bounds for neural networks
Testing Bayesian Networks
Further and stronger analogy between sampling and optimization: Langevin Monte Carlo and gradient descent
Non-Convex Learning via Stochastic Gradient Langevin Dynamics: A Nonasymptotic Analysis
Tight Bounds for Bandit Combinatorial Optimization
Greed Is Good: Near-Optimal Submodular Maximization via Greedy Optimization
Reliably Learning the ReLU in Polynomial Time
A Unified Analysis of Stochastic Optimization Methods Using Jump System Theory and Quadratic Constraints
Predicting with Distributions
Algorithmic Chaining and the Role of Partial Feedback in Online Nonparametric Learning
Towards Instance Optimal Bounds for Best Arm Identification
A General Characterization of the Statistical Query Complexity
Robust Sparse Estimation Tasks in High Dimensions*
Stochastic Composite Least-Squares Regression with convergence rate O(1/n)
Thresholding based Efficient Outlier Robust PCA
Effective Semisupervised Learning on Manifolds
Rates of estimation for determinantal point processes
High-Dimensional Regression with Binary Coefficients. Estimating Squared Error and a Phase Transition.
Online Learning Without Prior Information
Inapproximability of VC Dimension and Littlestone’s Dimension
Sparse Stochastic Bandits
Learning Non-Discriminatory Predictors
On Equivalence of Martingale Tail Bounds and Deterministic Regret Inequalities
Learning-Theoretic Foundations of Algorithm Configuration for Combinatorial Partitioning Problems
Learning with Limited Rounds of Adaptivity: Coin Tossing, Multi-Armed Bandits, and Ranking from Pairwise Comparisons
Robust Proper Learning for Mixtures of Gaussians via Systems of Polynomial Inequalities
Memory and Communication Efficient Distributed Stochastic Optimization with Minibatch Prox
Correspondence retrieval
Adaptivity to Noise Parameters in Nonparametric Active Learning
Matrix Completion from O(n) Samples in Linear Time
On Learning versus Refutation
Computationally Efficient Robust Estimation of Sparse Functionals*
Multi-Observation Elicitation
Square Hellinger Subadditivity for Bayesian Networks and its Applications to Identity Testing
Generalization for Adaptively-chosen Estimators via Stable Median
Thompson Sampling for the MNL-Bandit
Memoryless Sequences for Differentiable Losses
Fast rates for online learning in Linearly Solvable Markov Decision Processes
Efficient PAC Learning from the Crowd
Fast and robust tensor decomposition with applications to dictionary learning
Solving SDPs for synchronization and MaxCut problems via the Grothendieck inequality
Sample complexity of population recovery
On the Ability of Neural Nets to Express Distributions
Exact tensor completion with sum-of-squares
Nearly Optimal Sampling Algorithms for Combinatorial Pure Exploration
ZIGZAG: A new approach to adaptive online learning
*: Papers to be merged.