EPF5 - Week 1 - HackMD

# Introduction Hi EPFers, I am Ashley Yan. I recently graduated with a PhD in algebra and symbolic, with a focus on cryptography. I am interested in protocol design, verification, and the implementation of cryptographic schemes. Thanks for the opportunity to join EPF Cohort 5. It is a chance for me to dive deep into Ethereum core techniques and grow my expertise throughout this journey. # Process of Week 1 This week, I am in the discovery phase, exploring which topic or EIP to focus on. To better define my interests and the problem I want to address, I have been working on two main aspects. First, I am reviewing the research topics from the Ethereum protocol roadmap. Second, I am studying Vitalik's posts on Ethereum research from 2024. This approach is helping me align my interests with current and future developments in the Ethereum ecosystem. ## Research topics from Ethereum Protocol Roadmap Firstly, I went through the [Ethereum Protocol Roadmap](https://epf.wiki/#/wiki/research/roadmap), exploring the core R&D questions and future implementation topics. Several topics have caught my interest: ### 1. Formal verification in Ethereum: Safety and liveliness assurance is central to Ethereum's decentralized infrastructure. Formal verification plays a critical role in verifying correctness of: - The protocol's execution and consensus specifications. - Client implementations. - On-chain smart contract applications end users interact with. https://epf.wiki/#/wiki/testing/formal-verification One potential task is to find specific EIPs and conduct the verification process for them. For example, I could formally verify that EOF stack validation guarantees no stack underflows or overflows. ### 2. Single slot finality (SSF) to enable more validators I am researching [EIP-7514](https://eips.ethereum.org/EIPS/eip-7514), [EIP-7251](https://eips.ethereum.org/EIPS/eip-7251), and [8192 signature](https://ethresear.ch/t/sticking-to-8192-signatures-per-slot-post-ssf-how-and-why/17989), which aim to enable more validators. The technical challenge lies in efficiently coordinating an increased number of validators to achieve Secure Synchronized Functioning (SSF) with the best possible trade-offs. ## Recent posts by Vitalik on Ethereum Research - Jun 13: One-bit-per-attester inclusion lists https://ethresear.ch/t/one-bit-per-attester-inclusion-lists/19797 - May 11: MACI with mostly-off-chain “happy path” https://ethresear.ch/t/maci-with-mostly-off-chain-happy-path/19527 - April 19: A concrete proposal for correlated attester penalties https://ethresear.ch/t/a-concrete-proposal-for-correlated-attester-penalties/19341 - April 12: How to hard-fork to save most users’ funds in a quantum emergency https://ethresear.ch/t/how-to-hard-fork-to-save-most-users-funds-in-a-quantum-emergency/18901 - March 27: Supporting decentralized staking through more anti-correlation incentives https://ethresear.ch/t/supporting-decentralized-staking-through-more-anti-correlation-incentives/19116 - Dec: Sticking to 8192 signatures per slot post-SSF: how and why https://ethresear.ch/t/properties-of-issuance-level-consensus-incentives-and-variability-across-potential-reward-curves/18448 ### 1. One-bit-per-attester inclusion lists： The article discusses a technology called "inclusion lists" that distributes the authority to choose which transactions will be included in the next block. The goal is to decentralize this authority and prevent it from being concentrated in the hands of a single entity. Inclusion lists aim to provide a mechanism to improve the censorship resistance of Ethereum by allowing proposers to specify a set of transactions that must be promptly included for subsequent blocks to be considered valid, which is specified in EIP-7547. The proposed solution aims to decentralize the authority by spreading it across a highly decentralized set of actors, such as consensus block proposers. The authority to choose transactions is separated from the authority to order transactions, which tends to be more centralized due to economies of scale. Mechanism: - Seed Generation:When the block for slot N is published, a seed is generated from the RANDAO_REVEAL of the block. - Grouping:All attesters (validators) for that slot are divided into groups of size 2*T bytes, assuming each transaction is under T bytes (e.g., T = 500 bytes). The number of groups is k = attesters_per_slot / (2 * T). - Transaction Selection:Each attester is designated as the j’th attester of the i’th group. They identify the highest-priority-fee-paying valid transaction published before the slot N block, where hash(seed + tx) falls between 2256 / k * i and 2256 / k * (i+1). - Erasure Coding:The selected transaction is erasure-coded to 2T bits, and the attester publishes the j’th bit of this erasure encoding as part of their attestation. - Transaction Extraction:When these attestations are included in the next block, an algorithm like Berlekamp-Welch is used to attempt to reconstruct the transaction from the provided attester bits. ### 2. Supporting decentralized staking through more anti-correlation incentives This article explores a method to incentivize decentralization in blockchain protocols by penalizing correlated failures among validators. The idea is to impose greater penalties when multiple validators fail at the same time, which is more likely to happen with large stakers who control multiple validators. The research aims to encourage these stakers to decentralize their operations. Proposed Penalty Scheme： - Dynamic Penalty Calculation: In each slot, calculate 𝑝 p as the number of missed slots divided by the average number of misses over the last 32 slots, capped at 4. Penalties for missed attestations should be proportional to 𝑝 p. - Simulation of Penalty Schemes: Four penalty schemes were tested: Basic: Penalize one point per miss. Basic_ss: Same as basic but with single-slot inclusion. Excess: Penalize based on 𝑝 p. Excess_ss: Same as excess but with single-slot inclusion. - Results of Penalty Schemes:The excess penalty schemes reduce the advantage of large validators over small ones, making the system more fair. Large validators incur higher penalties for correlated failures, encouraging them to decentralize their infrastructure. # Plan of Week 2 1. Read the rest of schemes: - May 11: MACI with mostly-off-chain “happy path” https://ethresear.ch/t/maci-with-mostly-off-chain-happy-path/19527 - April 19: A concrete proposal for correlated attester penalties https://ethresear.ch/t/a-concrete-proposal-for-correlated-attester-penalties/19341 - April 12: How to hard-fork to save most users’ funds in a quantum emergency https://ethresear.ch/t/how-to-hard-fork-to-save-most-users-funds-in-a-quantum-emergency/18901 - Dec: Sticking to 8192 signatures per slot post-SSF: how and why https://ethresear.ch/t/properties-of-issuance-level-consensus-incentives-and-variability-across-potential-reward-curves/18448 2. Decide on the topic and research questions.