Timezone: »

Spotlight
Global Optimization Networks
Sen Zhao · Erez Louidor · Maya Gupta

Thu Jul 21 12:30 PM -- 12:35 PM (PDT) @ Room 310
We consider the problem of estimating a good maximizer of a black-box function given noisy examples. We propose to fit a new type of function called a global optimization network (GON), defined as any composition of an invertible function and a unimodal function, whose unique global maximizer can be inferred in $\mathcal{O}(D)$ time, and used as the estimate. As an example way to construct GON functions, and interesting in its own right, we give new results for specifying multi-dimensional unimodal functions using lattice models with linear inequality constraints. We extend to \emph{conditional} GONs that find a global maximizer conditioned on specified inputs of other dimensions. Experiments show the GON maximizers are statistically significantly better predictions than those produced by convex fits, GPR, or DNNs, and form more reasonable predictions for real-world problems.

#### Author Information

##### Maya Gupta (Univ. Washington)

Gupta is an Affiliate Professor of Electrical Engineering at the University of Washington, CEO and co-founder of the AI recommendations site Didero, CEO and co-founder of the AI distributed puzzle library Hoefnagel Puzzle Club, and CEO and founder of the laser-cut jigsaw puzzle manufacturer Artifact Puzzles. From 2013-2020, she led the Glassbox Machine Learning R&D team at Google Research, which focused on new ideas in constrained machine learning that were more accurate, interpretable, and fair. Her team's R&D had O(Billion) dollars of impact and improved O(Billiion) people's lives. Before joining Google, Gupta was an Associate Professor of Electrical Engineering at the University of Washington from 2003-2012, where she received the PECASE (presidential early career award for scientists and engineers), and Office of Naval Research Young Investigator Award for her work in statistical signal processing, and graduated 9 PhDs. Gupta received her PhD in EE from Stanford in 2003, and a BS EE and BA Economics from Rice University in 1997.