• Publications
  • Influence
Strong Duality for a Multiple-Good Monopolist
TLDR
We provide a duality-based framework for revenue maximization in a multiple-good monopoly and prove that the optimal values of these problems are always equal. Expand
  • 109
  • 14
  • PDF
Ten Steps of EM Suffice for Mixtures of Two Gaussians
TLDR
The Expectation-Maximization (EM) algorithm is a widely used method for maximum likelihood estimation in models with latent variables. Expand
  • 64
  • 11
  • PDF
Mechanism design via optimal transport
TLDR
We provide a framework for designing optimal mechanisms using optimal transport theory and duality theory to obtain conditions under which only pricing the grand bundle is optimal in multi-item settings. Expand
  • 80
  • 7
  • PDF
On the Power of Deterministic Mechanisms for Facility Location Games
TLDR
We consider K-Facility Location games, where n strategic agents report their locations in a metric space and a mechanism maps them to K facilities. Expand
  • 57
  • 6
  • PDF
The Complexity of Optimal Mechanism Design
TLDR
We show that it is hard to compute any encoding of an optimal auction of any format (direct or indirect, truthful or non-truthful) that can be implemented in expected polynomial time. Expand
  • 99
  • 5
  • PDF
On the Structure, Covering, and Learning of Poisson Multinomial Distributions
TLDR
An (n, k)-Poisson Multinomial Distribution (PMD) is the distribution of the sum of n independent random vectors supported on the set Bk={e1,...,ek} of standard basis vectors in Rk. Expand
  • 23
  • 5
  • PDF
A converse to Banach's fixed point theorem and its CLS-completeness
TLDR
We show that, whenever an iteration map globally converges to a unique fixed point, there exists a metric under which the iterative map is contracting and which can be used to bound the number of iterations until convergence. Expand
  • 21
  • 5
  • PDF
Learning Halfspaces with Massart Noise Under Structured Distributions
TLDR
We study the problem of learning halfspaces with Massart noise in the distribution-specific PAC model. Expand
  • 13
  • 3
  • PDF
A size-free CLT for poisson multinomials and its applications
TLDR
An (n,k)-Poisson Multinomial Distribution (PMD) is the distribution of the sum of n independent random vectors supported on the set Bk={e1,…,ek} of standard basis vectors in ℝk. Expand
  • 32
  • 2
  • PDF
Distribution-Independent PAC Learning of Halfspaces with Massart Noise
TLDR
We study the problem of {\em distribution-independent} PAC learning of halfspaces in the presence of Massart noise. Expand
  • 21
  • 2
  • PDF