块坐标算法描述,块坐标下降(BCD)方法由于迭代成本低、内存需求低、易于并行化以及能够发掘问题结构,在大规模数值优化中得到了广泛的应用。 Block coordinate descent (BCD) methods arewidely-used for large-scale numerical optimization because of their cheapiteration costs, low memory requirements, amenability to parallelization, andability to exploit problem structure.