论文信息 - Towards Interactive Feature Selection with Human-in-the-loop

Towards Interactive Feature Selection with Human-in-the-loop

Feature Selection (FS) has been applied to numerous domains, and shown to be effective in increasing the performance of machine learning algorithms. In the semiconductor industry, FS is part of various prediction tasks that aim at avoiding production stops and yield loss. For example, it can be used for: (i) diagnostics, wherein relevant features constitute potential root causes, with their identification being the initial step in a detailed investigation of process defects [3]; (ii) control, as the values of a small set of relevant features can be used to group objects and apply actions per group [1]; (iii) improving prediction performance and interpretability, by enforcing sparsity [4, 7]. Nevertheless, when analyzing manufacturing datasets, one faces two particular challenges, as in other real-world datasets:

[1] Daniel Hernández-Lobato,et al. Expectation propagation in linear regression models with spike-and-slab priors , 2015, Machine Learning.

[2] Richard van Haren,et al. Towards fab cycle time reduction by machine learning-based overlay metrology , 2018, European Mask and Lithography Conference.

[3] Richard van Haren,et al. Machine learning for fab automated diagnostics , 2017, European Mask and Lithography Conference.

[4] Julius Kunze,et al. Framework for Exploring and Understanding Multivariate Correlations , 2017, ECML/PKDD.

[5] Chenxi Lin,et al. A novel patterning control strategy based on real-time fingerprint recognition and adaptive wafer level scanner optimization , 2018, Advanced Lithography.

[6] Samuel Kaski,et al. Knowledge elicitation via sequential probabilistic inference for high-dimensional prediction , 2016, Machine Learning.

[7] Richard van Haren,et al. Pattern recognition and data mining techniques to identify factors in wafer processing and control determining overlay error , 2015, Advanced Lithography.