The absence of low-frequency (LF) component in seismic records leads to non-uniqueness in inversion, making the building of a reasonable LF model crucial for achieving high-precision acoustic impedance inversion. Artificial intelligence (AI) techniques represented by Gated Recurrent Unit (GRU) have shown favorable performance in the building of LF model. However, the problems of neural network overfitting and poor generalization due to small sample sizes remain challenging. We propose a neural network framework for LF model prediction towards small sample sizes. The framework utilizes the GRU as the basic unit and fully considers the long-term dependence in the time series of the LF model to ensure the integrity of vertical long-wavelength information. In order to improve the horizontal continuity of the LF model, a spatio-temporal feature fusion module is designed to convert the multi-trace data into potential vectors containing spatial information. Moreover, the seismic envelope and the initial model serve as prior information to guide the neural network, which is helpful for the network to yield stable predictions. In the synthetic and field data testing phase, spatial perturbations are introduced on the original data, it can generate more diverse labeled data. Experimental results demonstrate the predicted model on the basis of the initial model exhibit richer frequency components, more complete geological structures, and greater consistency with seismic records, and the predicted LF model of acoustic impedance not only aligns with the structural and depositional patterns of the area but also matches the measurements from a validation well.

Unsupervised network-based intrusion detection system (UNIDS) identifies suspicious traffic and alerts administrators without using any traffic labels. Existing Graph Convolutional Network (GCN)-based UNIDS approaches show great potential with collaboratively utilizing traffic features and network topologies. However, these methods suffer from an excessively high false-positive rate (FPR), e.g., 3.27% FPR in a supervised NIDS approach, while increases dramatically to 19.37% under the unsupervised setting. We reveal that the high FPR stems from a single-scale spatial-frequency learning paradigm, which blurs the distinction between benign and malicious traffic and misleads UNIDS systems. Therefore, we propose a Multi-scale Spatial-Frequency Intrusion Detection System (MSF-IDS) to mitigate the high FPR. Specifically, we propose multi-scale frequency encoders, thereby differentiating abnormal feature patterns. Then we propose NAPH as a spatial encoder, mining intrinsic abnormal topology patterns by tracking tens of thousands of evolving traffic nodes. To the best of our knowledge, NAPH takes the first step toward differentiable persistent homology analysis over dynamic network data. We also develop an executable application for NAPH to provide easy-access visualization insights. Finally, a self-supervised representation augmentor and an intrusion detector are proposed to refine and highlight the attack-specific information. Equipped with MSF-IDS, administrators effectively identify the unknown attack traffic, freeing security staff from labor-intensive engineering. Extensive experiments demonstrate the superiority of MSF-IDS, including binary classification, multi-classification, online intrusion detection, and visualized discussions. Our codes, datasets, and an executable application are available at https://github.com/qcydm/MSF-IDS.

Recent works in the domain of deep learning for object recognition on common image classification benchmarks often address the representation learning problem under the assumption of i.i.d. input data. Although achieving satisfying results, this assumption seems not realistic when agents have to learn autonomously. An autonomous agent receives a continual visual flow of objects which is far from an i.i.d. distribution of objects. Moreover, agents have to construct their representations of the world and adapt to unknown environments, without relying on external sources of information such as labels that would be provided post-classification and are unavoidable when an over-segmentation is done. Then, in order to exploit the learned representation effectively for object recognition, a clear and meaningful relationship w.r.t. real object categories is required, which has been largely neglected in existing unsupervised algorithms.In this paper, we propose a novelty detection method for continual and unsupervised object recognition, as an extension for the recent CURL model, which allows to moderate over-segmentation while preserving accuracy, in order to meet the requirements for autonomy. We experimentally validated our approach on two standard image classification benchmarks, MNIST and Fashion-MNIST, in this unsupervised and continual learning setting and improve the state of the art in terms of cluster purity, which is crucial for subsequent object recognition, since it facilitates clustering when information on ground truth labels is not available for free.