This blog post presents a novel risk-based approach to tune a deep model towards a target workload by its particular characteristics. Built on the recent advances on risk analysis for ER, the proposed approach first trains a deep model on labeled training data, and then fine-tunes it by minimizing its estimated misprediction risk on unlabeled target data. Our theoretical analysis shows that risk-based adaptive training can correct the label status of a mispredicted instance with a fairly good chance.
The proposed approach is shown in Figure 1, in which test data represent a target workload. It consists of two phases, the phase of traditional training followed by the phase of risk-based training. In the first phase, a deep model is trained on labeled training data in the traditional way. In the second phase, it is further tuned to minimize the misprediction risk on unlabeled target data.
The main contributions of this paper are as follows:
We propose a novel risk-based approach to enable adaptive deep learning.
We present a solution of adaptive deep learning for ER based on the proposed approach.
We theoretically analyze the performance of the proposed solution for ER. Our analysis shows that risk-based adaptive training can correct the label status of a mispredicted instance with a fairly good chance.
We empirically validate the efficacy of the proposed approach on real benchmark data by a comparative study.