Quantum advantage for learning shallow neural networks with natural data distributions

2025-03-26

AI Summary

This work addresses a central question in quantum machine learning: whether quantum algorithms can achieve provable learning advantages over classical methods for shallow periodic neurons under natural data distributions—such as Gaussian, generalized Gaussian, and logistic distributions. While prior results exhibit a binary dichotomy—exponential quantum speedups under uniform distributions versus no advantage under adversarial ones—this paper introduces the first efficient Quantum Statistical Query (QSQ) algorithm tailored to non-uniform natural distributions. Crucially, it extends quantum learning theory to real-valued classical function modeling, retaining strict exponential quantum advantage even under noise. Theoretically, the algorithm achieves polynomial quantum query complexity, whereas mainstream classical approaches—including gradient-based and statistical query methods—require exponential queries. This breakthrough transcends restrictive distributional assumptions, enabling practical applicability to real-world data.

Technology Category

• Machine Learning: Quantum Machine Learning
• Search and Optimization: Learning to Search
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

The application of quantum computers to machine learning tasks is an exciting potential direction to explore in search of quantum advantage. In the absence of large quantum computers to empirically evaluate performance, theoretical frameworks such as the quantum probably approximately correct (PAC) and quantum statistical query (QSQ) models have been proposed to study quantum algorithms for learning classical functions. Despite numerous works investigating quantum advantage in these models, we nevertheless only understand it at two extremes: either exponential quantum advantages for uniform input distributions or no advantage for potentially adversarial distributions. In this work, we study the gap between these two regimes by designing an efficient quantum algorithm for learning periodic neurons in the QSQ model over a broad range of non-uniform distributions, which includes Gaussian, generalized Gaussian, and logistic distributions. To our knowledge, our work is also the first result in quantum learning theory for classical functions that explicitly considers real-valued functions. Recent advances in classical learning theory prove that learning periodic neurons is hard for any classical gradient-based algorithm, giving us an exponential quantum advantage over such algorithms, which are the standard workhorses of machine learning. Moreover, in some parameter regimes, the problem remains hard for classical statistical query algorithms and even general classical algorithms learning under small amounts of noise.

Problem

• Quantum advantage in learning shallow neural networks
• Efficient quantum algorithm for non-uniform distributions
• Exponential quantum advantage over classical methods

Innovation

• Quantum algorithm for learning periodic neurons
• Exponential quantum advantage over classical methods
• Handles non-uniform data distributions effectively