LGAIOCNov 16, 2024

One-Layer Transformer Provably Learns One-Nearest Neighbor In Context

arXiv:2411.10830v121 citationsh-index: 16NIPS
Originality Incremental advance
AI Analysis

This provides theoretical insight into transformer mechanisms for researchers in machine learning, though it is incremental as it focuses on a specific, classical algorithm.

The paper tackles the problem of understanding how transformers perform in-context learning by proving that a single-layer transformer can learn the one-nearest neighbor classification rule from prompts, demonstrating this capability theoretically under nonconvex training conditions.

Transformers have achieved great success in recent years. Interestingly, transformers have shown particularly strong in-context learning capability -- even without fine-tuning, they are still able to solve unseen tasks well purely based on task-specific prompts. In this paper, we study the capability of one-layer transformers in learning one of the most classical nonparametric estimators, the one-nearest neighbor prediction rule. Under a theoretical framework where the prompt contains a sequence of labeled training data and unlabeled test data, we show that, although the loss function is nonconvex when trained with gradient descent, a single softmax attention layer can successfully learn to behave like a one-nearest neighbor classifier. Our result gives a concrete example of how transformers can be trained to implement nonparametric machine learning algorithms, and sheds light on the role of softmax attention in transformer models.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes