Adaptive Sparse Data Acquisition Technique Based on query-by-committee (QBC) and Sequential Spatial Adaptive Sampling(SSAS)

In this paper, an optimization method based on sequence space adaptive sampling (SSAS) and query by committee (QBC) is proposed to accelerate the near-field scanning of Integrated circuit, gradually reduce the sampling interval, and significantly reduce the number of scanned data samples. This method uses the query sub-committee method to collect data at high uncertainty locations. Combined with the SSAS algorithm, the sampling points with high uncertainty can be selected in batches, avoiding the repeated calculation of the sampling points with low uncertainty. The effectiveness of the proposed method is verified by numerical calculation and measurement results.