1 code implementation • 8 Feb 2020 • Ke Li, Zilin Xiang, Tao Chen, Shuo Wang, Kay Chen Tan
Given a tight computational budget, it is more cost-effective to focus on optimizing the parameter configuration of transfer learning algorithms (3) The research on CPDP is far from mature where it is "not difficult" to find a better alternative by making a combination of existing transfer learning and classification techniques.
no code implementations • 30 Jan 2019 • Ke Li, Zilin Xiang, Kay Chen Tan
Perhaps surprisingly, it is possible to build a cheap-to-evaluate surrogate that models the algorithm's empirical performance as a function of its parameters.