Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Sign up for free
Menu
Search
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Pricing
Search
Sign in
Sign up for free
Pathologies of Neural Models Make Interpretatio...
Search
Sponsored
·
Your Podcast. Everywhere. Effortlessly.
Share. Educate. Inspire. Entertain. You do you. We'll handle the rest.
→
Yasufumi Taniguchi
December 09, 2018
Research
1.8k
1
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
Pathologies of Neural Models Make Interpretations Difficult
Yasufumi Taniguchi
December 09, 2018
More Decks by Yasufumi Taniguchi
See All by Yasufumi Taniguchi
AllenNLPを使った開発
yasufumy
0
2.3k
Making Neural QA as Simple as Possible but not Simpler
yasufumy
0
110
Other Decks in Research
See All in Research
Les émissions de méthane entérique en élevage laitier : Enjeux et travaux menés en Afrique de l’Ouest et en France
institutdelelevage
PRO
0
130
Adapter les bergeries aux fortes chaleurs
institutdelelevage
PRO
0
140
Physical AIでモデリングはどう変わるか/How Physical AI Will Change Modeling
stktu
0
300
Karkada さんの論文 × 2 の紹介: (1) Closed-Form Training Dynamics Reveal Learned Features and Linear Structure in Word2Vec-like Models, (2) Symmetry in language statistics shapes the geometry of model representations
eumesy
PRO
1
860
SAM3を用いたコマ・吹き出しの 領域検出と分割構造からの読み順推定
kzmssk
0
190
Hitting the buffers
beale
0
280
某助成金プロジェクト採択に向けて企業研究所のアウトリーチ専任者がやったこと
afroscript
0
200
横浜市長(山中氏)の言動にかかる第三者による調査報告書
sishi2026
0
640
2025年度秋葉原ウォーカブルプロジェクト調査報告 「アキバらしいウォーカブル」とは何か
izumiyama_lab
1
250
GLIM とMegaParticles:正規分布近似の限界とタイトカップリング&パーティクルフィルタの進展 / GLIM and MegaParticles : Progress of the distribution representation in SLAM
koide3
0
960
Détecter les vulnérabilités dans un environnement souverain
hellosct1
0
110
学術バーQ AI研究最前線:自己教師あり学習による画像モデルの事前学習
naok615
0
170
Featured
See All Featured
For a Future-Friendly Web
brad_frost
184
10k
Navigating Team Friction
lara
192
16k
Future Trends and Review - Lecture 12 - Web Technologies (1019888BNR)
signer
PRO
0
3.8k
Lightning Talk: Beautiful Slides for Beginners
inesmontani
PRO
2
730
Fireside Chat
paigeccino
43
4.1k
Stewardship and Sustainability of Urban and Community Forests
pwiseman
0
550
Avoiding the “Bad Training, Faster” Trap in the Age of AI
tmiket
0
270
The Art of Delivering Value - GDevCon NA Keynote
reverentgeek
16
2.2k
How to build an LLM SEO readiness audit: a practical framework
nmsamuel
2
950
How to make the Groovebox
asonas
2
2.5k
The Web Performance Landscape in 2024 [PerfNow 2024]
tammyeverts
12
1.3k
Evolution of real-time – Irina Nazarova, EuRuKo, 2024
irinanazarova
9
1.6k
Transcript
ൃදऀ ୩ޱହ࢙ ҟৗͳڍಈ
!2 Pathological behavior ࣭จ͕did͚ͩͰ Ϟσϧͷग़ྗಉ͡ ֬ߴ͍
֓ཁ w NLPʹ͓͚ΔχϡʔϥϧϞσϧͷղੳख๏ΛఏҊ w Ϟσϧ͕λεΫΛղ্͘Ͱॏཁͳ୯ޠΛநग़͢Δख๏ w நग़͞Εͨ୯ޠਓʹͱͬͯҙຯෆ໌ w ҰํͰϞσϧநग़୯ޠͰਖ਼͘͠༧ଌ(Pathology) w
ղੳ݁Ռʹجͮ͘ਖ਼ଇԽ߲ΛఏҊ w ਖ਼ଇԽ߲ʹΑͬͯϞσϧͷղऍੑ্ !3
࣍ Ϟσϧղੳͷطଘख๏ ఏҊख๏ ࣮ݧ ·ͱΊ !4
Ϟσϧղੳͷطଘख๏
Ϟσϧղੳͷطଘख๏ !6 Adversarial Example Ϟσϧʹਓͷײʹ͢ΔڍಈΛͤ͞Δαϯϓϧ NLPͷλεΫ ओʹQAλεΫ Ͱύλʔϯ ਓʹͱͬͯҙຯͷͳ͍มߋ͕ɺϞσϧͷग़ྗΛܹมͤ͞Δέʔε
ਓʹͱͬͯ໌Β͔ͳมߋͰɺϞσϧ͕ग़ྗΛม͑ͳ͍έʔε
ग़ྗ͕ܹม͢Δέʔε !7 Jia et al., 2017 ΫΥʔλʔόοΫͷྸʹ͍ͭͯͷ จॻʹΫΥʔλʔόοΫͷഎ൪߸ʹ ؔ͢ΔจΛՃ Ϟσϧޡ
ग़ྗΛม͑ͳ͍έʔε !8 Mudrakarta et al., 2018 ݐͷന͍ϨϯΨ͕ରশ͔ʁ spherical (ٿঢ়ͷ) ݐͷന͍ϨϯΨ͕ٿঢ়͔ʁ
࣭จͷҙຯมԽ Ϟσϧͷ༧ଌෆม
2. ఏҊख๏
*OQVU3FEVDUJPO • ॏཁͰͳ͍୯ޠΛೖྗ͔ΒΓɺϞσϧͷڍಈΛੳ • Ϟσϧ͕ਖ਼͍͠ग़ྗΛ͢ΔͨΊʹඞཁͳ࠷୯ޠ (ॏཁ ୯ޠ) • Adversarial ExampleϞσϧʹͱͬͯͷॏཁ୯ޠʹண
*OQVU3FEVDUJPO !11 x y Ϟσϧͷ༧ଌ f( ⋅ ) χϡʔϥϧϞσϧ ೖྗܥྻ
(จจॻ) xi ೖྗܥྻͷ͋Δཁૉ (୯ޠ) g(xi |x) = f(y|x) − f(y|x−i ) ͋Δ୯ޠ ʹର͢Δ ॏཁΛఆٛ xi g i൪ͷ୯ޠΛফͨ͠ೖྗ
*OQVU3FEVDUJPO !12 g(xcontest |x) = f(y|x) − f(y|x−contest ) What
company won free advertisement due to QuickBooks contest ? What company won free advertisement due to QuickBooks contest ? g͕େ͖͚Εɺcontest͕ॏཁͳ୯ޠͱͳΔ Ϟσϧͷग़ྗʹେ͖͘د༩͍ͯ͠ΔͨΊ
*OQVU3FEVDUJPO !13 g(xi |x) = f(y|x) − f(y|x−i ) ॏཁͷ͍୯ޠΛআ
y͕มԽ͠ͳ͍Α͏ʹɺg͕࠷খͱͳΔ୯ޠiΛআ ͍ͯ͘͠
3. ࣮ݧ
ղੳͷରλεΫ 1. SQuAD w จॻͱ࣭จ͕༩͑ΒΕΔˠ࣭จʹରͯ͠Input Reduction w จॻ͔ΒղΛநग़͢ΔλεΫ 2. SNLI
w จ͕༩͑ΒΕΔˠͭͷจʹରͯ͠Input Reduction w จͷؔΛਪఆ͢ΔλεΫ 3. VQA w ը૾ͱ࣭จ͕༩͑ΒΕΔˠ࣭จʹରͯ͠Input Reduction w ղΛੜ͢ΔλεΫ !15
࣮ݧ༰ Input Reduction w Ϟσϧ͕ਖ਼͍͠ग़ྗΛ͢ΔαϯϓϧΛରʹ࣮ݧ w Input ReductionΛద༻ͨ͠ೖྗ(Reduced)ʹର͢ΔਓखධՁ w ReducedͱϥϯμϜʹ୯ޠΛམͱͨ͠߹(Random)ͷࠩҟͷධՁ
Regularization on Reduced Inputs w Input ReductionʹΑΔϞσϧͷPathological behaviorΛܰݮ͢Δਖ਼ଇԽ߲ ޙड़ ͷಋೖ !16
Reducedʹର͢ΔਓखධՁ !17 Reducedʹରͯ͠ ਓਖ਼͍͠༧ଌΛͰ ͖ͳ͍ w Reducedʹର͢Δਓͷਖ਼ w Ϟσϧͷਖ਼͕ͷαϯϓϧΛ༻
Reducedʹର͢ΔਓखධՁ !18 w ReducedͱRandomͷͲͪΒ͕ࣗવͳจ͔ w vs. Randomfifty-fiftyͱׂ͑ͨ߹ Reducedਓʹͱͬ ͯRandomͱಉ͡
Reducedͷࣄྫ !19 ʮͲ͜Ͱ࿅शͨ͠ ͔ʯΛฉ͔Ε͍ͯ ΔͷΘ͔Δ͕ɺ ʮͲͷνʔϜʯ͔ Θ͔Βͳ͍
Reducedͷฏۉ୯ޠ ͭͷλεΫͱɺ ਖ਼͢Δͷʹඞཁͳ୯ޠฏۉd
Reducedʹର͢ΔϞσϧͷ֬ !21 • Input Reductionͷద༻લޙͰϞσϧͷ ֬ʹมԽ΄ͱΜͲͳ͍ • ϞσϧӶ͍ϐʔΫΛ࣋ͭΑ͏ͳ Λֶश͍ͯ͠Δ͜ͱ͕ݪҼ
ਖ਼ଇԽ߲ͷಋೖ !22 ∑ (x,y)∈(X,Y) log(f(y|x)) + λ∑ ¯ x∈ ¯
X H(f(y| ¯ x)) Reducedʹରͯ͠ਖ਼͍͠yΛ ग़ྗ͠ʹ͘͘͢Δ ௨ৗͷతؔ Reducedαϯϓϧ௨ৗͷతؔΛֶͬͯशͨ͠ ϞσϧΛ༻͍ͯੜ
ਖ਼ଇԽ߲ͷޮՌ !23 • Ϟσϧͷਫ਼͕ඍ૿ • ਖ਼ʹඞཁͳ୯ޠ ͕૿Ճ
ਖ਼ଇԽ߲ͷޮՌ !24 ਓखධՁͷਫ਼্ Input Reductionͨ͠ೖྗ ͷղऍੑ্͕
ਖ਼ଇԽͨ͠Ϟσϧͷࣄྫ !25 Input Reductionͨ͠ೖྗ͕ਓͰ ղऍՄೳʹͳͬͨ
·ͱΊ ఏҊख๏ w NLPͷχϡʔϥϧϞσϧղੳख๏ͱͯ͠ɺInput ReductionΛఏҊ w ༧ଌʹد༩͠ͳ͍୯ޠΛೖྗ͔ΒΓɺϞσϧΛղੳ ࣮ݧ݁Ռ w ఏҊख๏Λద༻ͨ͠ೖྗਓʹͱͬͯҙຯෆ໌
w ҰํͰχϡʔϥϧϞσϧਖ਼͍͠༧ଌΛߦ͏ w ਖ਼ଇԽ߲Λಋೖ͢ΔͱϞσϧͷڍಈվળ !26