ML for identifying fraud using open blockchain data.pptx

Talk at ML IRL on 05/28/2024

1. ML for identifying fraud using open blockchain data Vijay Dialani Engineering Leader, Coinbase 28 May 2024 ‹#›

2. Vijay Dialani, PhD in Computer Science Applied ML Researcher with experience in Ads, Search, Recommendation Systems, Fraud Detection Systems and Crypto domains Prior experience in Privacy Preserving Machine Learning and Federated ML systems. Ex-Twitter, Ex-Google, Ex-Apple, Ex-LinkedIn, Ex-Microsoft Leading Machine Learning Risk and Machine Learning Platform teams at Coinbase 28 May 2024 ‹#›

3. Overview Introduction to Blockchain Common Attack Vectors Address Risk Scam detection for ERC20 Tokens Q＆A May 2024 ‹#›

4. May 2024 Introduction to Blockchain ML for identifying fraud using open blockchain data- Introduction to Blockchain ‹#›

5. What are some advantages of blockchains? Data security: helping to store data in a tamper-proof and immutable manner, with high availability using decentralized servers that are less susceptible to attack, manipulation, censorship and denial of service. Data provenance, traceability, auditability: recording transactions and assets in an immutable and transparent manner, facilitating tracking of the origin, ownership, provenance of data and agreements with digital signatures and time-stamping. This also provides an auditable and verifiable trail. Decentralized decision making: enabling decisions to be made by several entities or directly between two parties in settings where they do not have a prior existing trust-relationship among themselves or with a central entity. Autonomous and transparent code execution: enabling the execution of programs as smart contracts that are transparent to all concerned parties and operate autonomously without requiring trusted and centralized intermediaries. Decentralized identities: providing mechanisms for secure digital identities that allow users to interact with services without compromising privacy. Reference: https://www.coinbase.com/learn/crypto-basics/what-is-a-blockchain#how-blockchains-work Blockchains A blockchain is a list of transactions that anyone can view and verify. The Bitcoin blockchain, for example, contains a record of every time someone sent or received bitcoin. May 2024 ML for identifying fraud using open blockchain data - Introduction to Blockchains

6. Common Attack Vectors ‹#› May 2024 ML for identifying fraud using open blockchain data- Common Attack Vectors

7. Crypto Fraud Attack Vectors: Zero Transfer Phishing May 2024 ML for identifying fraud using open blockchain data - Common Attack Vectors Details: https://www.coinbase.com/blog/zero-transfer-phishing-part-1-attack-analysis

8. Crypto Fraud Attack Vectors: Pig Butchering Scams May 2024 ML for identifying fraud using open blockchain data - Common Attack Vectors Details: https://www.coinbase.com/blog/announcing-the-tech-against-scams-coalition The scam typically follows this chain of events: Victims are contacted through social media, are matched with the scammer on a dating app, or are contacted on instant messaging applications. The scammer encourages the victim to migrate their conversation to an encrypted messaging service, such as WhatsApp or WeChat, sometimes communicating for weeks or months before mentioning an investment opportunity. The scammer typically claims they have received great financial returns from a cryptocurrency trading or mining platform and convince their victim to co-invest with them or teach them how to trade successfully. Victims are directed to visit a fraudulent website that often looks like a legitimate trading platform and coached to deposit funds into an account that is controlled by the scammer. Some victims even receive a small amount of funds that are claimed to be “returns” on their investment to entice them to invest even larger sums. When the victim tries to withdraw funds from the site, they are then told they owe a tax payment or service fee before their funds will be released in an effort to further extort them for money.

9. Address Risk Indra Rustandi, Ayush Agarwal, Varsha Mahadevan, Vijay Dialani ‹#› May 2024 ML for identifying fraud using open blockchain data- Address Risk

10. May 2024 ML for identifying fraud using open blockchain data - Address Risk Details: https://www.coinbase.com/blog/detecting-fraudulent-transactions-coinbase-scalable-blockchain-address-risk Coinbase Scalable Blockchain Address Risk Scoring System

11. May 2024 ML for identifying fraud using open blockchain data - Address Risk Coinbase Scalable Blockchain Address Risk Scoring System Random walks generated for a graph with S as the starting point. Skip-gram based training strategy. node k represents the index for the kth-node in the graph.

12. Scam detection for ERC20 Tokens Yifan Xu, Yao Ma, Indra Rustandi, Vijay Dialani ‹#› May 2024 ML for identifying fraud using open blockchain data - Scam Detection for ERC20 Tokens

13. May 2024 Details: https://www.coinbase.com/blog/detecting-the-undetectable-coinbase-erc-20-scam-token-detection-system Coinbase ERC20 Scam Token Detection System ML for identifying fraud using open blockchain data - Scam Detection for ERC20 Tokens

14. Q＆A ‹#› May 2024 ML for identifying fraud using open blockchain data - Q＆A