Afterpre-processing methodsin [are 16,087 samples, and 1, 668, 738 featuresinthedataset.

Let B be the maximum difference betweenU1t and U2t, and let (π,θ1,θ2) be a Nash Equilibrium forG. Let π1 be the best response to the first teacher (with utilityU1t) and let π1+2 be the best response policy to the joint teacher. This result shows that as we reduce the number of random episodes, the approximation to aminimax regret strategy improves. Let G be the dual curriculum game in which the first teacher maximizes regret, so U1t = URt, and the second teacher plays randomly, soU2t = UUt . Finally,we need to show thatπ2+3 isoptimal for the student.

We investigate neural network representations from a probabilistic perspective.

Whilestochastic optimization ofAUROChasbeenstudied extensively, principled stochastic optimization of AUPRC has been rarely explored.

Conv-PCA is a naive method proposed in Section 4.1.

Yij. Assume now we have a graph embeddingZ: V Rd

Inversionoflinearsystems As mentioned in Corollary(1), for the Lasso, when computing the gradient off, one can either invert an nlinear system or anm mlinear system.