Lin Li

Haiyan Cai, Qingtang Jiang, Lin Li et al.

Recently the study of modeling a non-stationary signal as a superposition of amplitude and frequency-modulated Fourier-like oscillatory modes has been a very active research area. The synchrosqueezing transform (SST) is a powerful method for instantaneous frequency estimation and component separation of non-stationary multicomponent signals. The short-time Fourier transform-based SST (FSST for short) reassigns the frequency variable to sharpen the time-frequency representation and to separate the components of a multicomponent non-stationary signal. Very recently the FSST with a time-varying parameter, called the adaptive FSST, was introduced. The simulation experiments show that the adaptive FSST is very promising in instantaneous frequency estimation of the component of a multicomponent signal, and in accurate component recovery. However the theoretical analysis of the adaptive FSST has not been carried out. In this paper, we study the theoretical analysis of the adaptive FSST and obtain the error bounds for the instantaneous frequency estimation and component recovery with the adaptive FSST and the 2nd-order adaptive FSST.

Li Lin, Xinan He, Yan Ju et al.

Although effective deepfake detection models have been developed in recent years, recent studies have revealed that these models can result in unfair performance disparities among demographic groups, such as race and gender. This can lead to particular groups facing unfair targeting or exclusion from detection, potentially allowing misclassified deepfakes to manipulate public opinion and undermine trust in the model. The existing method for addressing this problem is providing a fair loss function. It shows good fairness performance for intra-domain evaluation but does not maintain fairness for cross-domain testing. This highlights the significance of fairness generalization in the fight against deepfakes. In this work, we propose the first method to address the fairness generalization problem in deepfake detection by simultaneously considering features, loss, and optimization aspects. Our method employs disentanglement learning to extract demographic and domain-agnostic forgery features, fusing them to encourage fair learning across a flattened loss landscape. Extensive experiments on prominent deepfake datasets demonstrate our method's effectiveness, surpassing state-of-the-art approaches in preserving fairness during cross-domain deepfake detection. The code is available at https://github.com/Purdue-M2/Fairness-Generalization