In the blockchain and open-source fields, efficient allocation of funds has always been a challenge. Nowadays, an innovative project called Deep Funding is trying to solve this issue with artificial intelligence and decentralized reviews. This project, supported by an initial funding of $250,000 from Vitalik Buterin, aims not only to address the resource allocation challenges currently facing the Ethereum ecosystem but also to pioneer new models for funding public goods in the future.
01. Deep Funding
What is Deep Funding?
Deep Funding is an innovative project that optimizes the allocation of funds for public goods through AI and decentralized review mechanisms, aiming to address the inefficiencies in resource allocation within the Ethereum ecosystem. The project's goal is to build a fair, transparent, and efficient funding allocation system to support Ethereum and its key open-source projects for long-term sustainable development.
Official website: https://deepfunding.org/
What problem do you want to solve?
Currently, there are the following problems with the allocation of funds for Ethereum public goods:
The irrationality of human decision-making: When faced with complex and abstract issues, humans often struggle to make reasonable judgments.
Preference for surface-level projects: Election-based funding mechanisms tend to favor funding projects that are superficially obvious, while neglecting deeper technical dependencies and complex contributions.
This has led to some crucial but 'hidden' infrastructure for the Ethereum ecosystem not receiving enough support, while also potentially wasting resources on projects that seem important in the short term but have limited long-term value.
What kind of thinking is used to solve problems?
The solutions proposed by Deep Funding include:
1. Constructing the Deep Graph
Deep Graph is a dynamic dependency graph that shows the relationships between projects and assigns weights to each dependency. In this way, the contributions and actual value of public goods can be visualized, solving the problem of measuring 'invisible contributions'.
2. AI model weighting and evaluation
Data input: Various information based on open-source projects (e.g., number of stars, contributor activity, last update, etc.). This requires you to use your imagination and understanding of the value of open-source projects.
Weight distribution: The AI model assigns weights based on the importance of dependencies and their actual impact, dynamically adjusting funding allocation.
Verification and optimization: The model undergoes spot checks by a jury to ensure the reasonableness of the weights.
3. Jury review mechanism
The jury is composed of experts who provide training data for the model by answering questions like 'Which is more important, Project A or Project B?'. This type of question was chosen because it is relatively easy for humans to discern and answer.
Collaboration model between humans and AI: Humans are responsible for direction and value judgments, while AI provides data analysis support. Subsequently, multiple models that closely align with human consensus will be selected for application.
4. Fair allocation of funds
Funds are allocated based on the contribution ratio of the projects, with a portion also incentivizing award-winning models.
Deep Funding will not only be used for weighting and distributing open-source software but this model can be applied to any scenario with dependencies and allocations. For example: papers, music, films, etc. Open-source software is just an initial attempt, and Deep Funding hopes to become a solution applicable to various scenarios.
02. Deep Funding Competition
Currently, the first competition of Deep Funding focuses on GitHub repos and open-source projects, constructing a weighted graph based on the dependencies of open-source projects to determine the donation amount each repository should receive. Then it focuses on open-source projects under the Ethereum label, especially clients.
The current progress of the Deep Funding project includes:
Sponsorship and funding: Vitalik Buterin provided an initial sponsorship of $250,000.
Data preparation: Collecting the Ethereum dependency graph, which involves data from about 40,000+ edges. This is already prepared.
Mechanism design: Conducting an AI model competition (to be held on Kaggle), currently recruiting AI models.
Trial evaluation: Validating the effectiveness of models through jury spot checks; applying the dependency weight models to Ethereum-related projects and observing the actual effects.
Of the $250K prize, $170k will be allocated to projects based on the weights of the dependency graph, $40k will reward the best-performing models in the review spot checks, and $40k will reward models submitted as open-source, with their innovation assessed by an expert jury.
There are still many challenges to face.
Fairness of reviews and incentive mechanisms: How to ensure the neutrality of the jury and the long-term participation enthusiasm? How to construct a fair and effective jury?
Effectiveness of AI models: How to accurately weight deep dependencies and avoid the model being misused or gamified?
Dynamic adjustment mechanisms: How to balance self-assessment and external review, avoiding bias?
Sources of funds and incentive methods: How to attract more funds to participate in the allocation, especially for non-code contributions?
We will proceed with discussions and explorations gradually.
