QFlowNet: Fast, Diverse, and Efficient Unitary Synthesis with Generative Flow Networks

2026-03-03

AI Summary

This work addresses key challenges in quantum circuit compilation—namely sparse rewards, low training efficiency, and insufficient solution diversity—by introducing generative flow networks (GFlowNets) to the task of unitary matrix decomposition for the first time. The proposed approach integrates a Transformer architecture to effectively learn from sparse reward signals and generate diverse quantum gate sequences proportional to their rewards. The Transformer encoder captures non-local structural properties of unitary matrices, enabling compact state representations, while the policy network facilitates efficient sampling. Evaluated on a 3-qubit benchmark with circuit depths ranging from 1 to 12, the method achieves a 99.7% success rate and discovers numerous compact, structurally diverse quantum circuits.

Technology Category

• Machine Learning: Quantum Machine Learning
• Search and Optimization: Learning to Search
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

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

Abstract

Unitary Synthesis, the decomposition of a unitary matrix into a sequence of quantum gates, is a fundamental challenge in quantum compilation. Prevailing reinforcement learning(RL) approaches are often hampered by sparse reward signals, which necessitate complex reward shaping or long training times, and typically converge to a single policy, lacking solution diversity. In this work, we propose QFlowNet, a novel framework that learns efficiently from sparse signals by pairing a Generative Flow Network (GFlowNet) with Transformers. Our approach addresses two key challenges. First, the GFlowNet framework is fundamentally designed to learn a diverse policy that samples solutions proportional to their reward, overcoming the single-solution limitation of RL while offering faster inference than other generative models like diffusion. Second, the Transformers act as a powerful encoder, capturing the non-local structure of unitary matrices and compressing a high-dimensional state into a dense latent representation for the policy network. Our agent achieves an overall success rate of 99.7% on a 3-qubit benchmark(lengths 1-12) and discovers a diverse set of compact circuits, establishing QFlowNet as an efficient and diverse paradigm for unitary synthesis.

Problem

• Unitary Synthesis
• Quantum Compilation
• Sparse Reward
• Solution Diversity
• Quantum Gates

Innovation

• Generative Flow Networks
• Unitary Synthesis
• Transformers
• Quantum Compilation
• Diverse Policy Learning