Glassoformer: A Query-Sparse Transformer for Post-Fault Power Grid Voltage Prediction
Glassoformer: A Query-Sparse Transformer for Post-Fault Power Grid Voltage Prediction
复制标题
DOI：
10.1109/icassp43922.2022.9747394
复制
发表时间：
2022-01
期刊：
影响因子：
--
通讯作者：
Yunling Zheng;Carson Hu;Guang Lin;Meng Yue;Bao Wang;Jack Xin
中科院分区：
文献类型：
--
作者：
Yunling Zheng;Carson Hu;Guang Lin;Meng Yue;Bao Wang;Jack Xin
我们提出了GLassoformer，一种新的和有效的Transformer架构，利用组Lasso正则化，以减少查询的标准自我注意力机制的数量。由于稀疏查询，GLassoformer比标准transformer计算效率更高。在电网故障后电压预测任务中，GLasso-former在准确性和稳定性方面明显优于许多现有的基准算法。
We propose GLassoformer, a novel and efficient transformer architecture leveraging group Lasso regularization to reduce the number of queries of the standard self-attention mechanism. Due to the sparsified queries, GLassoformer is more computationally efficient than the standard transformers. On the power grid post-fault voltage prediction task, GLasso-former shows remarkably better prediction than many existing benchmark algorithms in terms of accuracy and stability.