國立虎尾科技大學 |

Sequential decisions and predictions in natural language processing.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Sequential decisions and predictions in natural language processing./
作者:	He, He.
面頁冊數:	1 online resource (177 pages)
附註:	Source: Dissertation Abstracts International, Volume: 78-03(E), Section: B.
標題:	Artificial intelligence. -
電子資源:	click for full text (PQDT)
ISBN:	9781369140910

Sequential decisions and predictions in natural language processing.
He, He.

Sequential decisions and predictions in natural language processing. - 1 online resource (177 pages)

Source: Dissertation Abstracts International, Volume: 78-03(E), Section: B.

Thesis (Ph.D.)--University of Maryland, College Park, 2016.

Includes bibliographical references

Natural language processing has achieved great success in a wide range of applications, producing both commercial language services and open-source language tools. However, most methods take a static or batch approach, assuming that the model has all information it needs and makes a one-time prediction. In this dissertation, we study dynamic problems where the input comes in a sequence instead of all at once, and the output must be produced while the input is arriving. In these problems, predictions are often made based only on partial information. We see this dynamic setting in many real-time, interactive applications. These problems usually involve a trade-off between the amount of input received (cost) and the quality of the output prediction (accuracy). Therefore, the evaluation considers both objectives (e.g., plotting a Pareto curve).

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9781369140910Subjects--Topical Terms:

559380
Artificial intelligence.
Index Terms--Genre/Form:

554714
Electronic books.

Sequential decisions and predictions in natural language processing.
LDR:03976ntm a2200349K 4500 001 913656
005 20180622095235.5
006 m o u
007 cr mn||||a|a||
008 190606s2016 xx obm 000 0 eng d
020 $a 9781369140910
035 $a (MiAaPQ)AAI10159099
035 $a (MiAaPQ)umd:17384
035 $a AAI10159099
040 $a MiAaPQ $b eng $c MiAaPQ
100 1 $a He, He. $3 1186582
245 1 0 $a Sequential decisions and predictions in natural language processing.
264 0 $c 2016
300 $a 1 online resource (177 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: Dissertation Abstracts International, Volume: 78-03(E), Section: B.
500 $a Adviser: Hal Daume III.
502 $a Thesis (Ph.D.)--University of Maryland, College Park, 2016.
504 $a Includes bibliographical references
520 $a Natural language processing has achieved great success in a wide range of applications, producing both commercial language services and open-source language tools. However, most methods take a static or batch approach, assuming that the model has all information it needs and makes a one-time prediction. In this dissertation, we study dynamic problems where the input comes in a sequence instead of all at once, and the output must be produced while the input is arriving. In these problems, predictions are often made based only on partial information. We see this dynamic setting in many real-time, interactive applications. These problems usually involve a trade-off between the amount of input received (cost) and the quality of the output prediction (accuracy). Therefore, the evaluation considers both objectives (e.g., plotting a Pareto curve).
520 $a Our goal is to develop a formal understanding of sequential prediction and decision-making problems in natural language processing and to propose efficient solutions. Toward this end, we present meta-algorithms that take an existent batch model and produce a dynamic model to handle sequential inputs and outputs. We build our framework upon theories of Markov Decision Process (MDP), which allows learning to trade off competing objectives in a principled way. The main machine learning techniques we use are from imitation learning and reinforcement learning, and we advance current techniques to tackle problems arising in our settings. We evaluate our algorithm on a variety of applications, including dependency parsing, machine translation, and question answering. We show that our approach achieves a better cost-accuracy trade-off than the batch approach and heuristic-based decision-making approaches.
520 $a We first propose a general framework for cost-sensitive prediction, where different parts of the input come at different costs. We formulate a decision-making process that selects pieces of the input sequentially, and the selection is adaptive to each instance. Our approach is evaluated on both standard classification tasks and a structured prediction task (dependency parsing). We show that it achieves similar prediction quality to methods that use all input, while inducing a much smaller cost. Next, we extend the framework to problems where the input is revealed incrementally in a fixed order. We study two applications: simultaneous machine translation and quiz bowl (incremental text classification). We discuss challenges in this setting and show that adding domain knowledge eases the decision-making problem. A central theme throughout the chapters is an MDP formulation of a challenging problem with sequential input/output and trade-off decisions, accompanied by a learning algorithm that solves the MDP.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Artificial intelligence. $3 559380
650 4 $a Information technology. $3 559429
655 7 $a Electronic books. $2 local $3 554714
690 $a 0800
690 $a 0489
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of Maryland, College Park. $b Computer Science. $3 1180862
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10159099 $z click for full text (PQDT)