Data Flow

Here is the outline of the internal framework-

The SORACHAIN AI framework works by leveraging blockchain technology to host and train decentralized machine learning models in a transparent and accessible manner. Here is a breakdown of how the CMU framework operates based on the provided sources:

1. Decentralized Hosting and Training:

   • The CMU framework allows for the sharing and training of machine learning models in a decentralized manner. This means that the models are not controlled by a single entity but are accessible and updatable by a distributed network of participants.

2. Incentive Mechanism:

   • The framework includes an incentive mechanism that validates data contributions. Users may be required to provide a "stake" or deposit when adding data. This mechanism ensures the quality of the contributed data. Users can receive rewards for contributing good data or may face penalties for adding bad data.

3. Data Handling:

   • Data and metadata are stored on the blockchain by the Data Handler component. This ensures that the data is securely stored and accessible for future use beyond the scope of a single smart contract. 1

4. Machine Learning Model Updates:

   • The machine learning model is updated based on predefined training algorithms. As new data is contributed, the model is continuously improved and updated. Users can query the model for predictions, and the framework triggers the incentive mechanism to provide users with payments or virtual rewards.

5. Democratizing AI:

   • The overarching goal of the CMU framework is to democratize AI by encouraging decentralized hosting and versioning of public machine learning models. This aims to make AI more accessible and transparent, allowing anyone to contribute to model improvement.

6. Cost-Efficient Model Deployment:

   • Users can deploy models on the blockchain with a one-time deployment fee, which is typically a few dollars, instead of ongoing subscription fees to cloud service providers. This cost-effective approach enables broader access to machine learning models.

7. Encouraging Data Contribution:

   • Contributors can add data to the models in three steps: validation through the Incentive Mechanism, data storage by the Data Handler, and model updates based on training algorithms. The framework also provides various mechanisms to encourage contributors to submit high-quality data, such as gamification and self-assessment.

In essence, the SORACHAIN AI’s framework facilitates the collaborative improvement of machine learning models by enabling decentralized hosting, transparent data handling, and continuous model updates through blockchain technology, ultimately aiming to make AI more accessible and inclusive