We demonstrate the effectiveness of our approach on several real-world social networks.

We sincerely thank all of you for the detailed, thoughtful, and constructive comments and feedback. We added a table of racial composition data for all networks. We incorporated all the recommendations. We improve clarity of Th. 1 by adding "In this formulation, there are two sets of variables: a) We will provide a head-to-head comparison with Table 1. We will release the code and a "readme" file with instructions, detailing the sequence of the runs.

When using the robust performance metric described in Section 4.2, we have We solve the above linear program to obtain the results presented in Section 5.1. Work done while at UT Austin. We use Scipy's linear programming software (v 1.4.1) MDP is solved to obtain the sample's likelihood and determine the transition probabilities within the Markov chain. We used a learning rate of 0.01.

Neural Information Processing Systems http://nips.cc/

We develop parallel predictive entropy search (PPES), a novel algorithm for Bayesian optimization of expensive black-box objective functions. At each iteration, PPES aims to select a batch of points which will maximize the information gain about the global maximizer of the objective. Well known strategies exist for suggesting a single evaluation point based on previous observations, while far fewer are known for selecting batches of points to evaluate in parallel. The few batch selection schemes that have been studied all resort to greedy methods to compute an optimal batch. To the best of our knowledge, PPES is the first non-greedy batch Bayesian optimization strategy. We demonstrate the benefit of this approach in optimization performance on both synthetic and real world applications, including problems in machine learning, rocket science and robotics.

Bandit convex optimization is one of the fundamental problems in the field of online learning.

A.1 Proof for Theorem 1 A.1.1 Proof for (I) and (II) First, observe that the constraint in Equation ( 3) can be equivalently replaced by an inequality constraint f Therefore, the Lagrangian multiplier can be restricted to be λ 0. We have L II) follows a straightforward calculation. Proof for (III), the strong duality We first introduce the following lemma, which is a straight forward generalization of the strong Lagrange duality to functional optimization case. The proof of Lemma 1 is standard. However, for completeness, we include it here. Notice that both sets A and B are convex.

To this end, we prove a formula for the gradient of the exponential of matrices, which can be of practical interest on its own.