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
1.8k
Pathologies of Neural Models Make Interpretations Difficult
Yasufumi Taniguchi
December 09, 2018
Tweet
Share
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
98
Other Decks in Research
See All in Research
COFFEE-Japan PROJECT Impact Report(海ノ向こうコーヒー)
ontheslope
0
1.2k
YOLO26_ Key Architectural Enhancements and Performance Benchmarking for Real-Time Object Detection
satai
3
200
それ、チームの改善になってますか?ー「チームとは?」から始めた組織の実験ー
hirakawa51
0
1k
明日から使える!研究効率化ツール入門
matsui_528
10
5.7k
ペットのかわいい瞬間を撮影する オートシャッターAIアプリへの スマートラベリングの適用
mssmkmr
0
430
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
290
生成AI による論文執筆サポート・ワークショップ 論文執筆・推敲編 / Generative AI-Assisted Paper Writing Support Workshop: Drafting and Revision Edition
ks91
PRO
0
170
SREはサイバネティクスの夢をみるか? / Do SREs Dream of Cybernetics?
yuukit
3
450
Thirty Years of Progress in Speech Synthesis: A Personal Perspective on the Past, Present, and Future
ktokuda
0
200
SkySense V2: A Unified Foundation Model for Multi-modal Remote Sensing
satai
3
710
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
5
1.9k
第66回コンピュータビジョン勉強会@関東 Epona: Autoregressive Diffusion World Model for Autonomous Driving
kentosasaki
0
530
Featured
See All Featured
Practical Orchestrator
shlominoach
191
11k
Designing for Timeless Needs
cassininazir
0
180
Build your cross-platform service in a week with App Engine
jlugia
234
18k
[RailsConf 2023] Rails as a piece of cake
palkan
59
6.4k
No one is an island. Learnings from fostering a developers community.
thoeni
21
3.6k
Avoiding the “Bad Training, Faster” Trap in the Age of AI
tmiket
0
110
Are puppies a ranking factor?
jonoalderson
1
3.2k
The agentic SEO stack - context over prompts
schlessera
0
720
コードの90%をAIが書く世界で何が待っているのか / What awaits us in a world where 90% of the code is written by AI
rkaga
61
43k
Improving Core Web Vitals using Speculation Rules API
sergeychernyshev
21
1.4k
Building Adaptive Systems
keathley
44
3k
Game over? The fight for quality and originality in the time of robots
wayneb77
1
150
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