Abstract:The connectivity of networks has been widely studied in many high-impact applications, ranging from immunization, critical infrastructure analysis, social network mining, to bioinformatic system studies. Regardless of the end application domains, connectivity minimization has always been a fundamental task to effectively control the functioning of the underlying system. The combinatorial nature of the connectivity minimization problem imposes an exponential computational complexity to find the optimal solution, which is intractable in large systems. To tackle the computational barrier, greedy algorithm is extensively used to ensure a near-optimal solution by exploiting the diminishing returns property of the problem. Despite the empirical success, the theoretical and algorithmic challenges of the problems still remain wide open. On the theoretical side, the intrinsic hardness and the approximability of the general connectivity minimization problem are still unknown except for a few special cases. On the algorithmic side, existing algorithms are hard to balance between the optimization quality and computational efficiency. In this article, we address the two challenges by (1) proving that the general connectivity minimization problem is NP-hard and is the best approximation ratio for any polynomial algorithms, and (2) proposing the algorithm CONTAIN and its variant CONTAIN + that can well balance optimization effectiveness and computational efficiency for eigen-function based connectivity minimization problems in large networks.

Minimax regret 1-sink location problem with accessibility in dynamic general networks.

Minmax regret 1-sink location problems on dynamic flow path networks with parametric weights

Minmax Regret 1-Sink for Aggregate Evacuation Time on Path Networks

A Neural Network Model for General Minimax Problem

Zeroing-In On Network Metric Minima For Sink Location Determination

A new hybrid genetic algorithm for global minimax optimization

Sink Location Problems in Dynamic Flow Grid Networks

Minmax Tree Facility Location and Sink Evacuation with Dynamic Confluent Flows

Re-basin via implicit Sinkhorn differentiation

No-Regret Learning in Network Stochastic Zero-Sum Games

Minimax Regret Model for Liner Shipping Fleet Deployment with Uncertain Demand

On the Quickest Flow Problem in Dynamic Networks: a Parametric Min-Cost Flow Approach

On the Min-Max-Delay Problem: Complexity, Algorithm, and Int-Gap

Online distributed optimization with stochastic gradients: high probability bound of regrets

Minmax regret maximal covering location problems with edge demands

Online Distributed Optimization with Clipped Stochastic Gradients: High Probability Bound of Regrets

Optimal-transport satisficing with applications to capacitated hub location

A Sinkhorn-type Algorithm for Constrained Optimal Transport

Fast Connectivity Minimization on Large-Scale Networks

A MAXIMUM FLOW ALGORITHM IN SMALL CAPACITY NETWORKS