Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Pathologies of Neural Models Make Interpretatio...
Search
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
100
Other Decks in Research
See All in Research
【ローカルAIに向き合う展示会vol.2】液体時間定数型モジュールを用いた オリジナルの双方向エンコーダーモデルNexteraBERT 推論速度向上検討並びにダウンストリーム評価
rikkabotan7
0
140
多様なデータを許容し学習し続ける模倣学習 / Advanced Imitation Learning for VLA
prinlab
0
260
Dual Quadric表現を用いた動的物体追跡とRGB-D・IMU制約の密結合によるオドメトリ推定
nanoshimarobot
0
480
【中間報告】国会議員の立法・政策実務を支える環境を巡る現状と課題
polipoli
0
400
Language and AI
ayaniwa
0
190
人間中心の意思決定支援AI
yukinobaba
PRO
7
3.7k
言語モデルから言語について語る際に押さえておきたいこと
eumesy
PRO
7
2.5k
Research Engineerという仕事 / Research Engineering: Bridging Research and Business
chck
1
240
Vector Map as Language: Toward Unified Remote Sensing Vector Mapping
satai
3
140
【Zozo Research 技術共有会】三次元領域の現在と展望
mickey_0226
3
520
計算情報学研究室(数理情報学第7研究室)2026
tomohirokoana
0
700
第12回人と環境にやさしい交通をめざす全国大会/熊本都市圏「車1割削減、渋滞半減、公共交通2倍」をめざして
trafficbrain
0
140
Featured
See All Featured
Bridging the Design Gap: How Collaborative Modelling removes blockers to flow between stakeholders and teams @FastFlow conf
baasie
0
630
SERP Conf. Vienna - Web Accessibility: Optimizing for Inclusivity and SEO
sarafernandez
2
1.5k
How to Align SEO within the Product Triangle To Get Buy-In & Support - #RIMC
aleyda
2
1.8k
AI in Enterprises - Java and Open Source to the Rescue
ivargrimstad
0
1.4k
Understanding Cognitive Biases in Performance Measurement
bluesmoon
32
3k
Site-Speed That Sticks
csswizardry
13
1.4k
Reflections from 52 weeks, 52 projects
jeffersonlam
356
21k
Building Experiences: Design Systems, User Experience, and Full Site Editing
marktimemedia
0
560
Odyssey Design
rkendrick25
PRO
2
740
The Curious Case for Waylosing
cassininazir
1
450
The Anti-SEO Checklist Checklist. Pubcon Cyber Week
ryanjones
0
200
CSS Pre-Processors: Stylus, Less & Sass
bermonpainter
360
30k
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