國立虎尾科技大學 |

Concept-Monitor : = Using Concept Embeddings to Understand Neural Net Training.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Concept-Monitor :/
Reminder of title:	Using Concept Embeddings to Understand Neural Net Training.
Author:	Khan, Mohammad Ali.
Description:	1 online resource (51 pages)
Notes:	Source: Masters Abstracts International, Volume: 84-12.
Contained By:	Masters Abstracts International84-12.
Subject:	Computer engineering. -
Online resource:	click for full text (PQDT)
ISBN:	9798379761134

Concept-Monitor : = Using Concept Embeddings to Understand Neural Net Training.
Khan, Mohammad Ali.

Concept-Monitor :Using Concept Embeddings to Understand Neural Net Training. - 1 online resource (51 pages)

Source: Masters Abstracts International, Volume: 84-12.

Thesis (M.S.)--University of California, San Diego, 2023.

Includes bibliographical references

In this work, we propose a general framework called Concept-Monitor to help demystify the black-box DNN training processes automatically using a novel unified embedding space and concept diversity metric. Concept-Monitor enables human-interpretable visualization and indicators of the DNN training processes and facilitates transparency as well as deeper understanding on how DNNs develop along the during training. Inspired by these findings, we also propose a new training regularizer that incentivizes hidden neurons to learn diverse concepts, which we show to improve training performance. Finally, we apply Concept-Monitor to conduct several case studies on different training paradigms including adversarial training, fine-tuning and network pruning via the Lottery Ticket Hypothesis.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798379761134Subjects--Topical Terms:

569006
Computer engineering.
Subjects--Index Terms:

Explainable AIIndex Terms--Genre/Form:

554714
Electronic books.

Concept-Monitor : = Using Concept Embeddings to Understand Neural Net Training.
LDR:02151ntm a22003977 4500 001 1146240
005 20240812064349.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798379761134
035 $a (MiAaPQ)AAI30490746
035 $a AAI30490746
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Khan, Mohammad Ali. $3 1471598
245 1 0 $a Concept-Monitor : $b Using Concept Embeddings to Understand Neural Net Training.
264 0 $c 2023
300 $a 1 online resource (51 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: Masters Abstracts International, Volume: 84-12.
500 $a Advisor: Weng, Lily.
502 $a Thesis (M.S.)--University of California, San Diego, 2023.
504 $a Includes bibliographical references
520 $a In this work, we propose a general framework called Concept-Monitor to help demystify the black-box DNN training processes automatically using a novel unified embedding space and concept diversity metric. Concept-Monitor enables human-interpretable visualization and indicators of the DNN training processes and facilitates transparency as well as deeper understanding on how DNNs develop along the during training. Inspired by these findings, we also propose a new training regularizer that incentivizes hidden neurons to learn diverse concepts, which we show to improve training performance. Finally, we apply Concept-Monitor to conduct several case studies on different training paradigms including adversarial training, fine-tuning and network pruning via the Lottery Ticket Hypothesis.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Computer engineering. $3 569006
650 4 $a Computer science. $3 573171
653 $a Explainable AI
653 $a Interpretability
653 $a Machine learning
653 $a Concept-Monitor
653 $a DNN training processes
655 7 $a Electronic books. $2 local $3 554714
690 $a 0800
690 $a 0984
690 $a 0464
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of California, San Diego. $b Computer Science and Engineering. $3 1189479
773 0 $t Masters Abstracts International $g 84-12.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30490746 $z click for full text (PQDT)