Family learning: A process modeling method for cyber-additive manufacturing network
Citations
13,333 citations
11 citations
10 citations
4 citations
References
40,785 citations
"Family learning: A process modeling..." refers background or methods or result in this paper
...Moreover, if we model the individual product as one model for each, even with an advanced variable selection method, such as Lasso regression (Tibshirani, 1996), if the sample size is too small, then there will not be enough degrees of freedom to support model estimation with an accurate result....
[...]
...Both prediction performance and variable selection results outperform three benchmark methods, i.e., Lasso regression (Tibshirani, 1996), data-shared Lasso (Gross and Tibshirani, 2016) and MTL (Evgeniou and Pontil, 2004)....
[...]
...The results showed that the proposed family learning model outperforms Lasso regression (Tibshirani, 1996), data-shared Lasso (Gross and Tibshirani, 2016), and MTL (Evgeniou and Pontil, 2004), especially when the sample size is limited....
[...]
...…loss for model estimation; q1 is the tuning parameter which controls the sparsity of the model; the first penalty kBk1 ¼ P i bij j is a Lasso regularization (Tibshirani, 1996) term that forces the coefficients of insignificant variables to be zeros; q2 is the tuning parameter that…...
[...]
...The family learning is compared with three benchmark models to evaluate its prediction performance: (i) the Lasso regression (Tibshirani, 1996), which should have similar performance with the proposed method when the sample size is large enough; (ii) the data-shared Lasso (Gross and Tibshirani,…...
[...]
22,840 citations
18,616 citations
"Family learning: A process modeling..." refers background in this paper
...However, adequate samples from source and target domains are required to yield an accurate model for the target domain (Pan and Yang, 2010)....
[...]
16,538 citations
"Family learning: A process modeling..." refers background in this paper
...This can lead to unstable variable selection results with Lasso regularization terms (Zou and Hastie, 2005)....
[...]
13,333 citations
"Family learning: A process modeling..." refers methods in this paper
...As a physical experiment to physically quantify the quality performance of a SLM product is usually both time-consuming and economically expensive, a simulation study is employed (Montgomery, 2017)....
[...]
...A fractional factorial design (Montgomery, 2017) with three levels of settings and two kinds of designs is conducted to test the modeling performance....
[...]