The Cascade Algorithm: a New Approach for Distributed Markov-Based Optimization

This paper introduces a new stochastic optimisation approach in the form of a Cascade Optimisation Algorithm. The algorithm incorporates concepts from Markov processes whilst eliminating the inherent sequential nature that is a major obstacle preventing the exploitation of advances in distributed computing infrastructures. This method introduces partitions and pools to store intermediate solution and corresponding objectives. A Markov process increases the population of partitions and pools. The population is distributed periodically following an external certain.  With the use of partitions and pools, multiple Markov processes can be launched simultaneously for different partitions and pools. The cascade optimisation algorithm holds a potential in two different fronts. One aims at its deployment in parallel and distributed computing environments. Through storage of solutions in the pools, the algorithm further offers cost-effective means to analyze intermediate solutions, to visualize progress and to integrate optimization with data and/or knowledge management techniques without additional burden to the process.