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
OpenTalks.AI - Александр Чистяков, Построение о...
Search
OpenTalks.AI
February 15, 2019
Science
770
1
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
OpenTalks.AI - Александр Чистяков, Построение общего глубокого представления исполняемых файлов для поиска новых угроз и расследования киберпреступлений
OpenTalks.AI
February 15, 2019
Transcript
16 slides
1 Constructing shared deep representation of executable files to search
for new threats and cybercrime investigations Alexander Chistyakov Senior Research-Developer, Detection Methods Analysis, Kaspersky Lab
2 File processing route (known threat) Benign file Malicious file
Download file Check file’s reputation Static analysis Dynamic analysis Execution artefacts Raw file artefacts Status, popularity, sources, … Label obtained?
File processing route (modified known threat) Download file Check file’s
reputation Static analysis Dynamic analysis Execution artefacts Raw file artefacts Status, popularity, sources, … ML detection model Benign file Malicious file Label obtained?
File processing route (new unknown threat) Download file Check file’s
reputation Static analysis Dynamic analysis Execution artefacts Raw file artefacts Status, popularity, sources, … ML detection model Expert decision Benign file Malicious file
Manual data labeling Expert decision Benign file Malicious file Dynamic
analysis
6 World 2 Vec
Latent representations for malware Executable file 1. Polymorphic 2. Obfuscated
3. Selfpacked 4. Multicomponent Execution process 1. Context dependent 2. Unstable 3. Concurrent 4. Distributed
Evidence lower bound Variational Auto-Encoder (Basic) Variational Auto-Encoder (Symmetric)
File and file’s behavior joint distribution
File’s behavior conditional distribution
File’s and behavior shared embedding
Reducing internal traffic Expert decision Benign file Malicious file File’s
distribution approximator Dynamic analysis
Reducing external traffic User 1 User 2 User 3 Previously
observed malware collection File’s distribution approximator
Cybercrime investigations File’s distribution approximator Incident logs and artefacts Corporate
network Alarm!
What’s next? 1.Estimating real world file’s distribution 2.Avoiding model-based adversarial
attacks 3.Preventing private data leakage 4.Environment based anomaly detection
LET’S TALK? Kaspersky Lab HQ 39A/3 Leningradskoe Shosse Moscow, 125212,
Russian Federation Tel: +7 (495) 797-8700 www.kaspersky.com
More Decks by OpenTalks.AI
See All by OpenTalks.AI
OpenTalks.AI - Виктор Лемпицкий, Моделирование 3Д сцен: новые подходы в 2020 году
opentalks
0
510
OpenTalks.AI - Алексей Чернявский, Нейросетевые алгоритмы для повышения качества медицинских изображений
opentalks
0
460
OpenTalks.AI - Александр Громов, Устойчивость нейросетевых моделей при анализе КТ/НДКТ-исследований
opentalks
0
390
OpenTalks.AI - Денис Тимонин, Megatron-LM: Обучение мультимиллиардных LMs при помощи техники Model Parallelism
opentalks
0
550
OpenTalks.AI - Егор Филимонов, Возможности платформы Huawei Atlas и эффективный гетерогенный инференс.
opentalks
0
200
OpenTalks.AI - Александр Прозоров, Референсная архитектура робота сервисного центра в отраслях с изменчивыми бизнес-процессами
opentalks
0
410
OpenTalks.AI - Наталья Лукашевич, Анализ тональности по отношению к компании — с чем не справился BERT
opentalks
0
350
OpenTalks.AI - Константин Воронцов, Фейковые новости и другие типы потенциально опасного дискурса: типология, подходы, датасеты, соревнования
opentalks
0
470
OpenTalks.AI - Дмитрий Ветров, Фрактальность функции потерь, эффект двойного спуска и степенные законы в глубинном обучении - фрагменты одной мозаики
opentalks
0
510
Other Decks in Science
See All in Science
現代AI入門: 大規模言語モデルから思考するシステムへ — Test-Time Scaling, Loop, Harness, RSI
stillpedant
PRO
49
25k
データベース10: 拡張実体関連モデル
trycycle
PRO
0
1.6k
データベース08: 実体関連モデルとは?
trycycle
PRO
0
1.7k
Build your own LLM, Live, with MicroGPT
ianozsvald
0
180
[Relay webinaire eBIS actu n°16] Conclusion
institutdelelevage
PRO
0
320
明治薬科大学講義_ビッグデータ解析を支えるデータベース技術とクラウドコンピューティング
ktatsuya
1
220
1. CPC理論の展開と集合的知能モデル(JSAI2026 KS-27 集合的予測符号化と新たな知性の時代)
hayashiyus884
1
440
データベース09: 実体関連モデル上の一貫性制約
trycycle
PRO
0
1.9k
機械学習 - ニューラルネットワーク入門
trycycle
PRO
0
1.4k
データベース06: SQL (3/3) 副問い合わせ
trycycle
PRO
1
1.2k
[Relay webinaire eBIS actu n°16] Sélection pour la vitalité des jeunes bovins allaitants et laitiers
institutdelelevage
PRO
0
370
データベース14: B+木 & ハッシュ索引
trycycle
PRO
0
1k
Featured
See All Featured
WCS-LA-2024
lcolladotor
0
1k
Bootstrapping a Software Product
garrettdimon
PRO
306
120k
AI Search: Where Are We & What Can We Do About It?
aleyda
0
8.2k
Being A Developer After 40
akosma
PRO
92
590k
Primal Persuasion: How to Engage the Brain for Learning That Lasts
tmiket
0
520
Claude Code のすすめ
schroneko
67
230k
Building a Modern Day E-commerce SEO Strategy
aleyda
45
9.3k
Measuring Dark Social's Impact On Conversion and Attribution
stephenakadiri
2
330
How to build an LLM SEO readiness audit: a practical framework
nmsamuel
2
990
16th Malabo Montpellier Forum Presentation
akademiya2063
PRO
0
470
Let's Do A Bunch of Simple Stuff to Make Websites Faster
chriscoyier
508
140k
Digital Ethics as a Driver of Design Innovation
axbom
PRO
2
530