Source author record

Fangyu Li

Fangyu Li appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Computer Vision Cryptography and Security Distributed, Parallel, and Cluster Computing gr-qc Machine Learning

Catalog footprint

What is connected

3works

5topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2022arXiv

CPS Attack Detection under Limited Local Information in Cyber Security: A Multi-node Multi-class Classification Ensemble Approach

Cybersecurity breaches are the common anomalies for distributed cyber-physical systems (CPS). However, the cyber security breach classification is still a difficult problem, even using cutting-edge artificial intelligence (AI) approaches. In this paper, we study the multi-class classification problem in cyber security for attack detection. A challenging multi-node data-censoring case is considered. In such a case, data within each data center/node cannot be shared while the local data is incomplete. Particularly, local nodes contain only a part of the multiple classes. In order to train a global multi-class classifier without sharing the raw data across all nodes, the main result of our study is designing a multi-node multi-class classification ensemble approach. By gathering the estimated parameters of the binary classifiers and data densities from each local node, the missing information for each local node is completed to build the global multi-class classifier. Numerical experiments are given to validate the effectiveness of the proposed approach under the multi-node data-censoring case. Under such a case, we even show the out-performance of the proposed approach over the full-data approach.

preprint2020arXiv

Learning from Unlabelled Videos Using Contrastive Predictive Neural 3D Mapping

Predictive coding theories suggest that the brain learns by predicting observations at various levels of abstraction. One of the most basic prediction tasks is view prediction: how would a given scene look from an alternative viewpoint? Humans excel at this task. Our ability to imagine and fill in missing information is tightly coupled with perception: we feel as if we see the world in 3 dimensions, while in fact, information from only the front surface of the world hits our retinas. This paper explores the role of view prediction in the development of 3D visual recognition. We propose neural 3D mapping networks, which take as input 2.5D (color and depth) video streams captured by a moving camera, and lift them to stable 3D feature maps of the scene, by disentangling the scene content from the motion of the camera. The model also projects its 3D feature maps to novel viewpoints, to predict and match against target views. We propose contrastive prediction losses to replace the standard color regression loss, and show that this leads to better performance on complex photorealistic data. We show that the proposed model learns visual representations useful for (1) semi-supervised learning of 3D object detectors, and (2) unsupervised learning of 3D moving object detectors, by estimating the motion of the inferred 3D feature maps in videos of dynamic scenes. To the best of our knowledge, this is the first work that empirically shows view prediction to be a scalable self-supervised task beneficial to 3D object detection.

preprint2006arXiv

Perturbative photon flux generated by high-frequency relic gravitational waves and utilization of them for their detection

There exist corresponding metric perturbations of the relic gravitational waves (GWs) in the region of approximately h~10^(-30)-10^(-32)in the GHz band. A detector for these GWs is described in which we measure the perturbative photon flux (PPF) or signal generated by such high-frequency relic GWs (HFRGWs) via a coupling system of fractal membranes and a Gaussian beam (GB) passing through a static magnetic field. It is found that under the synchro-resonance condition in which the frequency of the GB is set equal to the frequency of the expected HFRGWs (h~2.00*10^(-31), v_g=10^10Hz in the quintessential inflationary models (QIM) and h~6.32*10^(-31), v_g=10^10Hz in the pre-big bang scenario (PBBS) may produce the PPFs of ~4.04*10^2/s and ~1.27*10^3/s in a surface of 100cm^2 area at the waist of the GB, respectively. The relatively weak first-order PPF, directed at right angles to the expected HFRGWs, is reflected by fractal membrane and the resulting reflected PPF (signal) exhibits a very small decay in transit to the detector (tunable microwave receiver) compared with the much stronger background photon flux, which allows for detection of the reflected PPF with signal to background noise ratios greater than one at the distance of the detector. We also discuss the selection capability of system and directional sensitivity for the resonance components from the stochastic relic GW background. The resolution of tiny difference between the PPFs generated by the relic GWs in the QIM and in the PBBS may be established and will be of cosmological significance. PACS numbers: 04.30.Nk, 04.30.Db, and 98.80.Cq.