Synthetic data can improve generalization when real data is scarce, but excessive reliance may introduce distributional mismatches that degrade performance. In this paper, we present a learning-theoretic framework to quantify the trade-off between synthetic and real data. Our approach leverages algorithmic stability to derive generalization error bounds, characterizing the optimal synthetic-to-real data ratio that minimizes expected test error as a function of the Wasserstein distance between the real and synthetic distributions. We motivate our framework in the setting of kernel ridge… Read More
Beyond Real Data: Synthetic Data through the Lens of Regularization
Synthetic data can improve generalization when real data is scarce, but excessive reliance may introduce distributional mismatches that degrade performance. In this paper, we present a learning-theoretic framework to quantify the trade-off between synthetic and real data. Our approach leverages algorithmic stability to derive generalization error bounds, characterizing the optimal synthetic-to-real data ratio that minimizes expected test error as a function of the Wasserstein distance between the real and synthetic distributions. We motivate our framework in the setting of kernel ridge…
Related Posts
AI Impact Summit 2026
A look at the partnerships and investments Google announced at the AI Impact Summit 2026.
MoEs Are Stronger than You Think: Hyper-Parallel Inference Scaling with RoE
The generation quality of large language models (LLMs) is often improved by utilizing inference-time sequence-level…
Learning to Evict from Key-Value Cache
The growing size of Large Language Models (LLMs) makes efficient inference challenging, primarily due to…