東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Deep Learning with Applications for ...

Wang, Jiachuan,

Linked to FindBook

Google Book

Amazon

博客來

Deep Learning with Applications for Spatiotemporal Prediction /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Deep Learning with Applications for Spatiotemporal Prediction // Jiachuan Wang.
Author:	Wang, Jiachuan,
Description:	1 electronic resource (131 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 86-05, Section: B.
Contained By:	Dissertations Abstracts International86-05B.
Subject:	Tin. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31718339
ISBN:	9798346542735

Deep Learning with Applications for Spatiotemporal Prediction /
Wang, Jiachuan,

Deep Learning with Applications for Spatiotemporal Prediction /Jiachuan Wang. - 1 electronic resource (131 pages)

Source: Dissertations Abstracts International, Volume: 86-05, Section: B.

Spatiotemporal prediction has garnered significant attention for many years. In recent years, deep learning methods have emerged as effective models for spatiotemporal data, surpassing traditional methods in tasks such as data enhancement and prediction. While considerable effort has been dedicated to developing deep learning methods for spa tiotemporal data mining, its popularity has soared due to the increasing need for high accuracy in managing vast amounts of data in the big data era. Besides, complex applica tion scenarios call for robust method designs that balance professionalism and versatility.This thesis aims to contribute methods to the pipeline of spatiotemporal prediction, which includes data enhancement as preprocessing and the design of effective tools for data modeling and prediction.In the first problem, our aim is to denoise noisy input frames such as radar images in situations where clean reference samples are not available, which is a common issue in spatiotemporal data collection. Constructing it as an image denoising problem, we use a self-supervised deep learning method that optimizes models based solely on noisyimages. The performance of this method heavily depends on the scale of the noise, which is often unknown. To address this issue, we closely estimate an upper bound for the noise scale without relying on clean images. Additionally, we design a framework for iterative model updates and scale estimation.For the secondproblem, weintroducepolynomialactivation functions to capture com plex and abstract high-order dynamics. This is particularly important for spatiotemporal data, such as rainfall systems, compared to simple image processing. Existing methods utilize activation functions to model non-linearity, but these functions have bounded first order derivatives. Taking inspiration from classical dynamic system simulation, we apply polynomial activation functions as a powerful tool for modeling high-order non-linearity. To further improve training robustness, we propose a Range Norm and prepend it before the activation functions. Besides, we develop an effective module to reduce monosemanticity, which aims to help the model acquire abstraction and reasoning abilities. Since large models designed for spatiotemporal prediction tend to learn inherent principles of dynamic systems slowly and expensively, we explore the variation of patterns across different scales. We hypothe size that the decrease in monosemanticity could be thereasonforthemodel'sgoodperfor mance. However, the decrease inmonosemanti city is passive, andaproactiveapproachto inducing it is still missing. To address this, we propose a framework that actively detects monosemantic neurons and inhibits them, with theoretical guarantees.Finally, extensive experiments on real and synthetic datasets demonstrate the effec tiveness of our proposed solutions.

English

ISBN: 9798346542735Subjects--Topical Terms:

3697305
Tin.

Deep Learning with Applications for Spatiotemporal Prediction /
LDR:04252nmm a22004093i 4500 001 2430566
005 20250527055043.5
006 m o d
007 cr|nu||||||||
008 261011s2024 miu||||||m |||||||eng d
020 $a 9798346542735
035 $a (MiAaPQD)AAI31718339
035 $a AAI31718339
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Wang, Jiachuan, $e author. $3 3818942
245 1 0 $a Deep Learning with Applications for Spatiotemporal Prediction / $c Jiachuan Wang.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (131 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 86-05, Section: B.
500 $a Advisors: Chen, Lei; Ng, Charles Wang Wai.
502 $b Ph.D. $c Hong Kong University of Science and Technology (Hong Kong) $d 2024.
520 $a Spatiotemporal prediction has garnered significant attention for many years. In recent years, deep learning methods have emerged as effective models for spatiotemporal data, surpassing traditional methods in tasks such as data enhancement and prediction. While considerable effort has been dedicated to developing deep learning methods for spa tiotemporal data mining, its popularity has soared due to the increasing need for high accuracy in managing vast amounts of data in the big data era. Besides, complex applica tion scenarios call for robust method designs that balance professionalism and versatility.This thesis aims to contribute methods to the pipeline of spatiotemporal prediction, which includes data enhancement as preprocessing and the design of effective tools for data modeling and prediction.In the first problem, our aim is to denoise noisy input frames such as radar images in situations where clean reference samples are not available, which is a common issue in spatiotemporal data collection. Constructing it as an image denoising problem, we use a self-supervised deep learning method that optimizes models based solely on noisyimages. The performance of this method heavily depends on the scale of the noise, which is often unknown. To address this issue, we closely estimate an upper bound for the noise scale without relying on clean images. Additionally, we design a framework for iterative model updates and scale estimation.For the secondproblem, weintroducepolynomialactivation functions to capture com plex and abstract high-order dynamics. This is particularly important for spatiotemporal data, such as rainfall systems, compared to simple image processing. Existing methods utilize activation functions to model non-linearity, but these functions have bounded first order derivatives. Taking inspiration from classical dynamic system simulation, we apply polynomial activation functions as a powerful tool for modeling high-order non-linearity. To further improve training robustness, we propose a Range Norm and prepend it before the activation functions. Besides, we develop an effective module to reduce monosemanticity, which aims to help the model acquire abstraction and reasoning abilities. Since large models designed for spatiotemporal prediction tend to learn inherent principles of dynamic systems slowly and expensively, we explore the variation of patterns across different scales. We hypothe size that the decrease in monosemanticity could be thereasonforthemodel'sgoodperfor mance. However, the decrease inmonosemanti city is passive, andaproactiveapproachto inducing it is still missing. To address this, we propose a framework that actively detects monosemantic neurons and inhibits them, with theoretical guarantees.Finally, extensive experiments on real and synthetic datasets demonstrate the effec tiveness of our proposed solutions.
546 $a English
590 $a School code: 1223
650 4 $a Tin. $3 3697305
650 4 $a Neurons. $3 588699
650 4 $a Precipitation. $3 3680799
650 4 $a Deep learning. $3 3554982
650 4 $a Back propagation. $3 3681810
650 4 $a Neural networks. $3 677449
650 4 $a Data processing. $3 680224
650 4 $a Design. $3 518875
650 4 $a Weather forecasting. $3 546213
650 4 $a Natural language processing. $3 1073412
650 4 $a Polynomials. $3 604754
650 4 $a Rain. $3 3560372
650 4 $a Atmospheric sciences. $3 3168354
650 4 $a Mathematics. $3 515831
650 4 $a Meteorology. $3 542822
690 $a 0389
690 $a 0800
690 $a 0725
690 $a 0405
690 $a 0557
710 2 $a Hong Kong University of Science and Technology (Hong Kong). $e degree granting institution. $3 3814653
720 1 $a Chen, Lei $e degree supervisor.
720 1 $a Ng, Charles Wang Wai $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-05B.
790 $a 1223
791 $a Ph.D.
792 $a 2024
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31718339