Upgrade to Pro — share decks privately, control downloads, hide ads and more …

CAPTCHAとボットの共進化

Sponsored · Ship Features Fearlessly Turn features on and off without deploys. Used by thousands of Ruby developers. →
Avatar for Yuma Kurogome Yuma Kurogome
April 26, 2016
Technology
1.2k
2

 CAPTCHAとボットの共進化

http://connpass.com/event/29969/ LT資料

Avatar for Yuma Kurogome

Yuma Kurogome

April 26, 2016

More Decks by Yuma Kurogome

See All by Yuma Kurogome
The Art of De-obfuscation
Avatar for Yuma Kurogome ntddk
16
28k
死にゆくアンチウイルスへの祈り
Avatar for Yuma Kurogome ntddk
55
39k
Windows Subsystem for Linux Internals
Avatar for Yuma Kurogome ntddk
10
3.1k
なぜマルウェア解析は自動化できないのか
Avatar for Yuma Kurogome ntddk
6
4.4k
Linear Obfuscation to Drive angr Angry
Avatar for Yuma Kurogome ntddk
4
910
マルウェアを機械学習する前に
Avatar for Yuma Kurogome ntddk
3
1.7k
Peeling Onions
Avatar for Yuma Kurogome ntddk
7
3.8k
仮想化技術を用いたマルウェア解析
Avatar for Yuma Kurogome ntddk
8
27k
An Introduction to Drawbridge(ja)
Avatar for Yuma Kurogome ntddk
11
3.5k

Other Decks in Technology

See All in Technology
Purview 勉強会報告 Microsoft Purview 入門しようとしてみた
Avatar for masakichi masakichixo
1
450
なぜ、IAMロールのプリンシパルに*による部分マッチングが使えないのか? / 20260518-ssmjp-iam-role-principal
Avatar for opelab opelab
1
130
社内RAGの導入で気を付けたポイント
Avatar for やくも yakumo
1
130
Terragrunt x Snowflake + dbt で作るマルチテナントなデータ基盤構築プラットフォーム
Avatar for Gaku TASHIRO gak_t12
0
460
Swift Sequence の便利 API 再発見
Avatar for treastrain / Tanaka Ryoga treastrain
1
290
続 運用改善、不都合な真実 〜 物理制約のない運用改善はほとんど無価値 / 20260518-ssmjp-kaizen-no-value-without-physical-constraints
Avatar for opelab opelab
2
250
20260515 ログイン機能だけではないアカウント管理を全体で考える～サービス設計者向け～
Avatar for OpenID Foundation Japan oidfj
1
770
おいらのAWSアップデートの追い方〜Slack×AgentCore〜
Avatar for やくも yakumo
1
110
業務に残された「良くない型」で考える「TypeScriptの難しさ」
Avatar for Saji sajikix
2
460
ECSのTerraformモジュールにコントリビュートした話
Avatar for Haruka Sakihara harukasakihara
0
240
JaSSTに関わることで変わった人生観 #jasstnano
Avatar for Makky makky_tyuyan
0
140
既存プロダクトQAから新規プロダクトQAへ
Avatar for hashi ryotakahashi
0
160

Featured

See All Featured
What’s in a name? Adding method to the madness
Avatar for The Alliance productmarketing
PRO
24
4k
How to Create Impact in a Changing Tech Landscape [PerfNow 2023]
Avatar for Tammy Everts tammyeverts
55
3.3k
How GitHub (no longer) Works
Avatar for Zach Holman holman
316
150k
Designing for Timeless Needs
Avatar for Cassini Nazir cassininazir
1
220
Building a A Zero-Code AI SEO Workflow
Avatar for Ian Lurie portentint
PRO
0
520
JavaScript: Past, Present, and Future - NDC Porto 2020
Avatar for David Neal reverentgeek
52
5.9k
HU Berlin: Industrial-Strength Natural Language Processing with spaCy and Prodigy
Avatar for Ines Montani inesmontani
PRO
0
380
Templates, Plugins, & Blocks: Oh My! Creating the theme that thinks of everything
Avatar for Michelle Schulp Hunt marktimemedia
31
2.8k
Fireside Chat
Avatar for paigeccino paigeccino
42
3.9k
How People are Using Generative and Agentic AI to Supercharge Their Products, Projects, Services and Value Streams Today
Avatar for Helen Beal helenjbeal
1
180
Breaking role norms: Why Content Design is so much more than writing copy - Taylor Woolridge
Avatar for UX Y'all uxyall
0
290
Agile Actions for Facilitating Distributed Teams - ADO2019
Avatar for Mark Kilby mkilby
0
190

Transcript

  1. None

  2. • •

  3. None

  4. •

  5. None

  6. • •

  7. None
  8. None
  9. None

  10. def __init__(self, input_channel, output_channel, filter_height, filter_width, mid_units, n_units, n_label): super(CNN,

    self).__init__( conv1 = L.Convolution2D(input_channel, output_channel, (filter_height, filter_width)), l1 = L.Linear(mid_units, n_units), l2 = L.Linear(n_units, n_label), ) • • • • • • •