Databricks has created a novel machine learning technique that significantly improves AI model performance, addressing the widespread issue of dirty data. Jonathan Frankle, the company's chief AI scientist, discusses how many businesses lack clean data for effective model fine-tuning. The new method uses reinforcement learning combined with synthetic data to enable companies to deploy AI agents for various tasks without data quality hindrances. It highlights the importance of 'best-of-N' strategies for enhancing model effectiveness, which is becoming central to modern AI development.
"Everybody has some data, and has an idea of what they want to do... But the lack of clean data makes it challenging to fine-tune a model to perform a specific task."
"The method leverages ideas that have helped produce advanced reasoning models by combining reinforcement learning, a way for AI models to improve through practice, with 'synthetic,' or AI-generated training data."
Collection
[
|
...
]