skip to main content

Search for: All records

Creators/Authors contains: "Liu, S"

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. Free, publicly-accessible full text available June 1, 2023
  2. Free, publicly-accessible full text available January 1, 2023
  3. Free, publicly-accessible full text available January 1, 2023
  4. Free, publicly-accessible full text available February 7, 2023
  5. Reliable probability estimation is of crucial importance in many real-world applications where there is inherent (aleatoric) uncertainty. Probability-estimation models are trained on observed outcomes (e.g. whether it has rained or not, or whether a patient has died or not), because the ground-truth probabilities of the events of interest are typically unknown. The problem is therefore analogous to binary classification, with the difference that the objective is to estimate probabilities rather than predicting the specific outcome. This work investigates probability estimation from high-dimensional data using deep neural networks. There exist several methods to improve the probabilities generated by these models but they mostly focus on model (epistemic) uncertainty. For problems with inherent uncertainty, it is challenging to evaluate performance without access to ground-truth probabilities. To address this, we build a synthetic dataset to study and compare different computable metrics. We evaluate existing methods on the synthetic data as well as on three real-world probability estimation tasks, all of which involve inherent uncertainty: precipitation forecasting from radar images, predicting cancer patient survival from histopathology images, and predicting car crashes from dashcam videos. We also give a theoretical analysis of a model for high-dimensional probability estimation which reproduces several of the phenomena evincedmore »in our experiments. Finally, we propose a new method for probability estimation using neural networks, which modifies the training process to promote output probabilities that are consistent with empirical probabilities computed from the data. The method outperforms existing approaches on most metrics on the simulated as well as real-world data.« less
    Free, publicly-accessible full text available January 1, 2023
  6. Simultaneous human activities, such as the Super Bowl game, would cause certain impacts on frequency fluctuations in power systems. With the help of FNET/GridEye measurements, this paper aims to give comprehensive analyses on the frequency fluctuations during Super Bowl LIV held on Feb. 2, 2020, so as to better understand several phenomena caused by simultaneous activities which will help system operations and controls. First, recent developments of the FNET/GridEye are briefly introduced. Second, the frequency fluctuations of the Eastern Interconnection (EI), western electricity coordinating council (WECC), and electric reliability council of Texas (ERCOT) power systems during Super Bowl LIV are analyzed. Third, frequency fluctuations of Super Bowl Sunday and ordinary Sundays in 2020 are compared. Finally, the differences of frequency fluctuations among different years during the Super Bowl and their change trends are also given. Furthermore, several possible explanations, including the simultaneity of electricity consumption at the beginning of commercial breaks and the halftime show, the increasing usage of the Internet, and the increasing size of TV screens, are illustrated in detail in this paper.