相关论文: When Outliers Are Different

Fast radio bursts (FRBs) are a type of highly-polarized, millisecond-duration electromagnetic pulses in the radio band, which are mostly produced at cosmological distances. These properties provide a natural laboratory for testing the…

Currently, FRB 121102 is the only fast radio burst source that was observed to give out bursts repeatedly. It shows a high repeating rate, with more than one hundred bursts being spotted, but with no obvious periodicity in the activities.…

Machine learning and data analysis have been used in many robotics fields, especially for modelling. Data are usually the result of sensor measurements and, as such, they might be subjected to noise and outliers. The presence of outliers…

Gamma Ray Bursts (GRBs) are notorious for their diversity. Yet, they have a series of common features. The typical energy of their $\gamma$ rays is a fraction of an MeV. The energy distributions are well described by a ``Band spectrum'',…

Gamma ray bursts (GRBs) are short and intense pulses of $\gamma$-rays arriving from random directions in the sky. Several years ago Amelino-Camelia et al. pointed out that a comparison of time of arrival of photons at different energies…

To investigate the repeater content of gamma ray burst samples I develop two models where sources burst at a constant average rate. I find that the sky coverage affects the number of repeaters in a sample predominantly through the detector…

A large dimensional characterization of robust M-estimators of covariance (or scatter) is provided under the assumption that the dataset comprises independent (essentially Gaussian) legitimate samples as well as arbitrary deterministic…

Fast Radio Bursts (FRBs) are millisecond-duration transient events that are typically observed at radio wavelengths and cosmological distances but their origin remains unclear. Furthermore, most FRB origin models are related to the…

This dissertation investigates the use of one-sided classification algorithms in the application of separating hazardous chlorinated solvents from other materials, based on their Raman spectra. The experimentation is carried out using a new…

Functional data analysis can be seriously impaired by abnormal observations, which can be classified as either magnitude or shape outliers based on their way of deviating from the bulk of data. Identifying magnitude outliers is relatively…

Multi-dimensional scaling (MDS) plays a central role in data-exploration, dimensionality reduction and visualization. State-of-the-art MDS algorithms are not robust to outliers, yielding significant errors in the embedding even when only a…

A new method for outlier detection and generation is introduced by lifting data into the space of probability distributions which are not analytically expressible, but from which samples can be drawn using a neural generator. Given a…

A scenario is proposed that explains both the observed high pulsar velocities and extragalactic gamma-ray bursts (GRBs). The model involves an ultra- relativistic jet from a supernova (SN), that produces a GRB and its afterglow, whose…

We consider whether the variability properties of Gamma-Ray Bursts (GRBs) that produce bright optical and longer wavelength transient afterglows (A-GRBs) are the same as a larger, inclusive sample of bright, long-duration GRBs, selected…

Outliers are ubiquitous in modern data sets. Distance-based techniques are a popular non-parametric approach to outlier detection as they require no prior assumptions on the data generating distribution and are simple to implement. Scaling…

Spatial perception is the backbone of many robotics applications, and spans a broad range of research problems, including localization and mapping, point cloud alignment, and relative pose estimation from camera images. Robust spatial…

Universal outlier hypothesis testing refers to a hypothesis testing problem where one observes a large number of length-$n$ sequences -- the majority of which are distributed according to the typical distribution $\pi$ and a small number…

We study the spectro-temporal characteristics of two repeating fast radio bursts (FRBs), namely, FRB 20180916B and FRB 20180814A, and combine the results with those from our earlier analysis on FRB 20121102A. The relationship between the…

We propose an inlier-based outlier detection method capable of both identifying the outliers and explaining why they are outliers, by identifying the outlier-specific features. Specifically, we employ an inlier-based outlier detection…

Gamma-ray bursts (GRB) are extreme events. They are crudely classified into two groups based on their duration, namely the short and long bursts. Such a classification has proven to be useful to determine their progenitors: the merger of…