“AISTATS is an exciting conference that allows for engaging conversations and interactions at the intersection of machine learning and statistics. ML@GT is thrilled to be a part of this growing conference and we are looking forward to connecting with other researchers from around the world,” said Sebastian Pokutta, associate director of ML@GT and a paper author.

ML@GT faculty members Le Song, Byron Boots, and Negar Kiyavash are 2019 area chairs.

Georgia Tech’s twelve papers:

• Nearly Optimal Adaptive Procedure for Piecewise-Stationary Bandit: a Change-Point Detection Approach
• Robust Submodular Maximization: Offline and Online Algorithms
• Restarting Frank-Wolfe
• Truncated Back-propagation for Bilevel Optimization
• Accelerating Imitation Learning with Predictive Models
• Risk-Averse Stochastic Convex Bandit
• Differentially Private Online Submodular Minimization
• Structured Robust Submodular Maximization: Offline and Online Algorithms
• Kernel Exponential Family Estimation via Doubly Dual Embedding
• Stochastic Variance-Reduced Cubic Regularization for Nonconvex Optimization
• Database Alignment with Gaussian Features
• On Landscape of Lagrangian Function for Stochastic Search for Constrained Nonconvex Optimization