1·I had to define the task closure, determine appropriate task granularity, split the subproblem, combine the results, and so on.
我必须定义任务闭包、确定合适的任务粒度、分离子问题、综合结果，等等。
2·The process will often be self-repeating since each subproblem may still be complex enough to require further decomposition.
由于每个子问题可能仍然十分复杂，需要进一步的分解，这个过程就将不断的循环往复。
3·Chapter 2: a trust region subproblem model with memory is proposed.
第二章：给出了带记忆的信赖域子问题模型。
4·The subproblem is solved by simulated annealing algorithm.
该子问题可通过模拟退火算法来解决。