• Publications
  • Influence
Multiresolution Matrix Factorization
TLDR
This paper introduces a new notion of matrix factorization that can capture structure in matrices at multiple different scales. Expand
  • 40
  • 6
  • PDF
Online optimization for the smart (micro) grid
TLDR
We extend recent work in the field of online optimization to the scheduling of generators in smart (micro) grids and derive bounds on the performance of asymptotically good algorithms in terms of the generator parameters. Expand
  • 47
  • 6
  • PDF
Adaptivity to Local Smoothness and Dimension in Kernel Regression
TLDR
We present the first result for kernel regression where the procedure adapts locally at a point x to both the unknown local dimension of the metric space X and the unknown Holder-continuity of the regression function at x. Expand
  • 34
  • 6
  • PDF
Novel Biobjective Clustering (BiGC) Based on Cooperative Game Theory
TLDR
We propose a new approach to clustering based on cooperative game theory. Expand
  • 29
  • 3
  • PDF
Generalization and Representational Limits of Graph Neural Networks
TLDR
We address two fundamental questions about graph neural networks (GNNs). Expand
  • 39
  • 3
  • PDF
DEEP-CARVING: Discovering visual attributes by carving deep neural nets
TLDR
We propose Deep-Carving, a novel training procedure with CNNs, that helps the net efficiently carve itselffor the task of multiple attribute prediction. Expand
  • 57
  • 2
  • PDF
Link Label Prediction in Signed Social Networks
TLDR
We introduce a novel problem which we call the link label prediction problem: Given the information about signs of certain links in a social network, we want to learn the nature of relationships that exist among the users by predicting the sign, positive or negative, of the remaining links. Expand
  • 43
  • 2
  • PDF
Generative Models for Graph-Based Protein Design
TLDR
We develop relational language models for protein sequences that directly condition on a graph specification of the target structure. Expand
  • 41
  • 1
  • PDF
Highly scalable parallel collaborative filtering algorithm
TLDR
In this paper, we present the design of a soft real-time (around 1 min.) parallel CF algorithm based on the Concept Decomposition technique and demonstrate the performance and scalability of our algorithm. Expand
  • 12
  • 1
Learning Tree Structured Potential Games
TLDR
We focus on learning tree structured potential games where equilibria are represented by local maxima of an underlying potential function. Expand
  • 12
  • 1
  • PDF
...
1
2
3
4
...