TITAN: A Trajectory-Informed Technique for Adaptive Parameter Freezing in Large-Scale VQE

2025-09-18

AI Summary

Large-scale variational quantum eigensolver (VQE) training faces two critical bottlenecks: high gradient evaluation overhead—scaling linearly with the number of parameters—and slow convergence due to barren plateaus, causing exponential growth in measurement cost. To address these challenges, this work introduces TITAN, a novel framework that dynamically predicts parameter importance via trajectory-based analysis during training. TITAN integrates barren-plateau-resilient data construction with an adaptive neural architecture to intelligently identify and freeze inactive parameters early in optimization. Crucially, it requires no additional quantum resources and generalizes across diverse ansätze. On 30-qubit systems, TITAN reduces circuit evaluations by 40–60% and accelerates convergence threefold, while preserving or improving ground-state energy estimation accuracy. These advances significantly enhance the scalability of VQE for quantum chemistry and materials simulation.

Technology Category

• Machine Learning: Quantum Machine Learning
• Search and Optimization: Learning to Search
• Computer Vision: Learning & Optimization for CV

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Variational quantum Eigensolver (VQE) is a leading candidate for harnessing quantum computers to advance quantum chemistry and materials simulations, yet its training efficiency deteriorates rapidly for large Hamiltonians. Two issues underlie this bottleneck: (i) the no-cloning theorem imposes a linear growth in circuit evaluations with the number of parameters per gradient step; and (ii) deeper circuits encounter barren plateaus (BPs), leading to exponentially increasing measurement overheads. To address these challenges, here we propose a deep learning framework, dubbed Titan, which identifies and freezes inactive parameters of a given ansatze at initialization for a specific class of Hamiltonians, reducing the optimization overhead without sacrificing accuracy. The motivation of Titan starts with our empirical findings that a subset of parameters consistently has a negligible influence on training dynamics. Its design combines a theoretically grounded data construction strategy, ensuring each training example is informative and BP-resilient, with an adaptive neural architecture that generalizes across ansatze of varying sizes. Across benchmark transverse-field Ising models, Heisenberg models, and multiple molecule systems up to 30 qubits, Titan achieves up to 3 times faster convergence and 40% to 60% fewer circuit evaluations than state-of-the-art baselines, while matching or surpassing their estimation accuracy. By proactively trimming parameter space, Titan lowers hardware demands and offers a scalable path toward utilizing VQE to advance practical quantum chemistry and materials science.

Problem

• Reduces VQE training overhead for large Hamiltonians
• Addresses barren plateaus and measurement overheads
• Freezes inactive parameters without sacrificing accuracy

Innovation

• Adaptive parameter freezing technique
• Deep learning framework for VQE
• BP-resilient data construction strategy