Search

The generative autoencoders, such as the variational autoencoders or the adversarial autoencoders, have achieved great success in lots of real-world applications, including image generation and signal communication. However, little concern has been devoted to their robustness during practical deployment. Due to the probabilistic latent structure, variational autoencoders (VAEs) may confront problems such as a mismatch between the posterior distribution of the latent and real data manifold, or discontinuity in the posterior distribution of the latent. This leaves a back door for malicious attackers to collapse VAEs from the latent space, especially in scenarios where the encoder and decoder are used separately, such as communication and compressed sensing. In this work, we provide the first study on the adversarial robustness of generative autoencoders in the latent space. Specifically, we empirically demonstrate the latent vulnerability of popular generative autoencoders through attacks in the latent space. We also evaluate the difference between variational autoencoders and their deterministic variants and observe that the latter performs better in latent robustness. Meanwhile, we identify a potential trade-off between the adversarial robustness and the degree of the disentanglement of the latent codes. Additionally, we also verify the feasibility of improvement for the latent robustness of generative autoencoders through adversarial training. In summary, we suggest concerning the adversarial latent robustness of the generative autoencoders, analyze several robustness-relative issues, and give some insights into a series of key challenges. [ABSTRACT FROM AUTHOR]

The normalized subband adaptive filtering algorithm provides fast convergence rate for colored input signals as compared to the normalized least mean square algorithm, but it suffers from a poor convergence issue in the α -stable noise. In light of this, the normalized subband p-norm (NSPN) algorithm, which is based on the least mean p-power error (MPE) criterion, is proposed in this study. This technique is not only robust against impulsive noise samples, but it also maintains a fast convergence rate when colored input signals are used. In addition to this, we develop both the steady-state and the transient models of the NSPN algorithm and provide some insights. Then, in order to solve the problem of making a choice regarding the order p in the NSPN algorithm, we design an autonomous system and come up with the NSPN algorithm with a variable p-norm (VP-NSPN). In addition, we offer the TFC-based VP-NSPN algorithm with a fast convergence rate and low steady-state misadjustment simultaneously by making use of the tap-weights feedback-based convex combination (TFC) scheme. In conclusion, simulation results on system identification and acoustic echo cancellation are undertaken in order to validate the superiority of the proposed algorithms and validate the usefulness of the theoretical analysis. [ABSTRACT FROM AUTHOR]

Identification of Hammerstein systems with polynomial features and mean square error loss has received a lot of attention due to their simplicity in calculation and solid theoretical foundation. However, when the prior information of nonlinear subblock of a Hammerstein system is unknown or some outliers are involved, the performance of related methods may degenerate seriously. The main reason is that the used polynomial just has finite approximation capability to an unknown nonlinear function, and mean square error loss is sensitive to outliers. In this paper, a new identification method based on random Fourier features and kernel risk sensitive loss is therefore proposed. Since the linear combination of random Fourier features can well approximate any continuous nonlinear function, it is expected to be more powerful to characterize the nonlinear behavior of Hammerstein systems. Moreover, since the kernel risk sensitive loss is a similarity measure that is insensitive to outliers, it is expected to have excellent robustness. Based on the mean square convergence analysis, a sufficient condition to ensure the convergence and some theoretical values regarding the steady-state excess mean square error of the proposed method are also provided. Simulation results on the tasks of Hammerstein system identification and electroencephalogram noise removal show that the new method can outperform other popular and competitive methods in terms of accuracy and robustness. [ABSTRACT FROM AUTHOR]

Using lower limb rehabilitation robots (LLRRs) to help stroke patients recover their walking ability is attracting more and more attention presently. Previous studies have shown that gait rehabilitation training with natural gait pattern can improve the therapeutic outputs. However, how to generate the personalized gait trajectory has not been well researched. In this paper, a personalized gait generation method based anthropometric features is proposed. Firstly, gait trajectories are fitted and simplified into Fourier coefficient vectors, which are used to represent gait trajectories. Secondly, fourteen body features are used to generate the personalized gait trajectories and the feature set is further optimized based on the minimal redundancy maximal relevance criterion for easy application on the LLRR. Then, the relationship between the optimized feature set and gait trajectories is modeled by using the RF algorithm. Finally, the performance of the proposed method is demonstrated by several comparison experiments. [ABSTRACT FROM AUTHOR]

Copyright of Frontiers of Information Technology & Electronic Engineering is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Accumulating evidence suggests that the brain is highly dynamic; thus, investigation of brain dynamics especially in brain connectivity would provide crucial information that stationary functional connectivity could miss. This study investigated temporal expressions of spatial modes within the default mode network (DMN), salience network (SN) and cognitive control network (CCN) using a reliable data-driven co-activation pattern (CAP) analysis in two independent data sets. We found enhanced CAP-to-CAP transitions of the SN in patients with MDD. Results suggested enhanced flexibility of this network in the patients. By contrast, we also found reduced spatial consistency and persistence of the DMN in the patients, indicating reduced variability and stability in individuals with MDD. In addition, the patients were characterized by prominent activation of mPFC. Moreover, further correlation analysis revealed that persistence and transitions of RCCN were associated with the severity of depression. Our findings suggest that functional connectivity in the patients may not be simply attenuated or potentiated, but just alternating faster or slower among more complex patterns. The aberrant temporal–spatial complexity of intrinsic fluctuations reflects functional diaschisis of resting-state networks as characteristic of patients with MDD. [ABSTRACT FROM AUTHOR]