Optimization
A framework for bilevel optimization that enables stochastic and global variance reduction algorithms
Figure 1: Convergence curves of the two proposed methods on a toy problem.
Learning the Efficient Frontier
This optimal solution is traditionally found by solving a convex optimization problem.
Symbolic Regression via Neural-Guided Genetic Programming Population Seeding
On a number of common benchmark tasks to recover underlying expressions from a dataset, our method recovers 65% more expressions than a recently published top-performing model using the same experimental setup.
Supplementary Material: Efficient hierarchical Bayesian inference for spatio-temporal regression models in neuroimaging Summary of the proposed algorithms and derived update rules
Choose a random initial value for p as well as h, and construct H = diag( h) and P = diag( p) .
Piper: Multidimensional Planner for DNN Parallelization
In the "modern era", such model-parallel training techniques trace their roots back to AlexNet [