SimQFL: A Quantum Federated Learning Simulator with Real-Time Visualization
Ratun Rahman 0 , Atit Pokharel 0 , Md Raihan Uddin 0 , and Dinh C. Nguyen 0 · 2025
Read SimQFL: A Quantum Federated Learning Simulator with Real-Time Visualization by Ratun Rahman 0 , Atit Pokharel 0 , Md Raihan Uddin 0 , and Dinh C. Nguyen 0 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-12477.txt.
Introduction
Ratun Rahman, Atit Pokharel, Md Raihan Uddin, and Dinh Nguyen built a simulator for quantum federated learning in a 2025 paper. Most quantum simulators target circuits alone, leaving training loops and live feedback to hand code. Their tool was designed from the start for learning across many clients.
After each training round the screen updates with fresh results and moving learning curves, so users can watch convergence or spot a broken circuit early. Knobs cover rounds, learning rate, client count, qubit number, and circuit depth. Users can also load their own data, a gap in many rivals. The authors pitch the platform as a friendly bench for prototyping quantum neural networks. Students and small teams are the intended first users.
Catalog record
| CREATOR | Ratun Rahman 0 , Atit Pokharel 0 , Md Raihan Uddin 0 , and Dinh C. Nguyen 0 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-12477
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
SimQFL: A Quantum Federated Learning Simulator with Real-Time Visualization Ratun Rahman 0 , Atit Pokharel 0 , Md Raihan Uddin 0 , and Dinh C. Nguyen 0 Abstract Quantum federated learning (QFL) is an emerging field that has the potential to revolutionize computation by taking advantage of quantum physics concepts in a distributed machine learning (ML) environment. However, the majority of available quantum simulators are primarily built for general quantum circuit simulation and do not include integrated support for machine learning tasks such as training, evaluation, and iterative optimization. Furthermore, designing and assessing quantum learning algorithms is still a difficult and resource-intensive task. Real-time updates are essential for observing model convergence, debugging quantum circuits, and making conscious choices during training with the use of limited resources. Furthermore, most current simulators fail to support the integration of user-specific data for training purposes, undermining the main purpose of using a simulator. In this study, we introduce SimQFL, a customized simulator that simplifies and accelerates QFL experiments in quantum network applications. SimQFL supports real-time, epoch-wise output development and visualization, allowing researchers to monitor the process of learning across each training round. Furthermore, SimQFL offers an intuitive and visually appealing interface that facilitates ease of use and seamless execution. Users can customize key variables such as the number of epochs, learning rates, number of clients, and quantum hyperparameters such as qubits and quantum layers, making the simulator suitable for various QFL applications. The system gives immediate feedback following each epoch by showing intermediate outcomes and dynamically illustrating learning curves. User can also upload their own data and check the simulation results using QFL without any fundamental knowledge. SimQFL is a practical and interactive platform enabling academics and developers to prototype, analyze, and tune quantum neural networks with greater transparency and control in distributed quantum networks. Emails: [email protected], [email protected], [email protected], [email protected] I Introduction Simulators are important in both conventional and quantum computing research because they enable algorithm development, testing, and validation in controlled and repeatable environments. In the field of federated learning (FL), simulators allow academics to experiment with collaborative training methodologies without the requirement for a large-scale distributed infrastructure. Open-source FL platforms, such as OpenFL [ 1 ] and PeerFL [ 2 ] , have emerged as important tools, supporting a wide range of training scenarios and facilitating rigorous evaluation of federated algorithms across a variety of network, client, and data settings. Similarly, simulators such as Fujitsu’s 40-qubit system [ 3 ] and platforms described at QSim 2024 [ 4 ] are critical environments for developing quantum circuits and evaluating quantum algorithms on classical hardware. These simulators not only decrease the barrier to testing but also provide insights into algorithm behavior, resource efficiency, and performance bottlenecks. Quantum machine learning (QML) investigates the integration of quantum computing principles with machine learning approaches in order to provide computational benefits such as faster processing, richer data representations, and improved model performance [ 5 ] . Using quantum features such as superposition and entanglement, QML has brought new paradigms in data encoding, neural topologies, and optimization. However, one major problem with QML is that i) it requires data centralization , which creates data privacy concerns [ 6 ] , and ii) centralized quantum processing , which requires significant quantum resources at the central server [ 7 ] . By addressing these drawbacks with decentralized, privacy-preserving learning processes called federated learning (FL), quantum federated learning (QFL) shows significant potential as an extension to QML. Without disclosing raw data or needing centralized quantum processing, QFL allows many quantum clients to train local quantum models on confidential datasets and collaboratively update a shared global model [ 8 , 9 ] . This configuration offers built-in benefits in data security, scalability, and error tolerance while significantly reducing the requirement for centralized quantum hardware. By modeling diverse client behaviors, multiple data distributions (including non-IID settings), and varied quantum configurations (such as the number of qubits or circuit depth), a well-designed QFL simulator can facilitate experimentation with these decentralized quantum models [ 10 ] . Such a simulator helps researchers evaluate performance trade-offs between privacy, accuracy, and utilization of resources, in addition to assisting with algorithmic development. However, current simulators exhibit two practical challenges. i) Lacking integrated support for QFL: QFL platforms lack quantum features like qubit encoding, variational quantum layers, or support for hybrid quantum-classical models, despite their great adaptability in conventional settings [ 11 ] . In contrast, general-purpose quantum simulators emphasize circuit reliability and quantum gates but provide minimal support for multi-client synchronization, model training, or optimization [ 12 ] . As a result, researchers studying QFL are frequently obliged to improvise tools or experiment with non-interactive scripts. This brings up the first important question: a) How can researchers use only classical hardware to efficiently build and examine the training dynamics of QML models in a federated environment? ii) Providing a visually appealing interface and real-time insight: In addition, current simulators are inadequate in terms of ease of execution and visually appealing during training. Subsequently, conventional FL or quantum simulation tools hardly enable real-time visualization, which is crucial for managing computing resources, detecting convergence, and troubleshooting quantum layers. For example, methods that investigate non-IID data impacts in QFL, such as the one by Zhao et al. [ 13 ] , depend on post-training assessments without ongoing observation. Comparably, privacy-preserving QFL frameworks that emphasize cryptographic integration, as MQFL by Dutta et al. [ 14 ] , lack visual interpretability during training. This raises a second crucial question: b) How can researchers use visually appealing, real-time insights into performance and learning behavior to examine and refine QFL models interactively? iii) Personal data training: Furthermore, most existing QFL and simulation systems do not allow users to upload and use their own datasets for personalized model training. This shortcoming greatly affects the usefulness of such tools, especially when the reason for using a simulator is to apply a learning method without fundamental knowledge. In order to create reliable and broadly applicable QFL systems, simulators must be able to incorporate user-specific data. Otherwise, they are unable to represent client-specific distributions, preferences, or noise characteristics. As a result, this raises a third key question: c) In order to facilitate user-centric experimentation in QFL research, how can simulators facilitate the smooth and private integration of personal datasets? Main Contributions: To answer these significant questions, we introduce SimQFL, a simulator for quantum federated learning that has integrated support for real-time visualization and distributed quantum model training . The primary goal is to provide a useful, adaptable, and interactive platform that lets developers and researchers test, prototype, and evaluate QFL algorithms without requiring complicated distributed systems or quantum tools. This paper’s main contributions are listed below. • We present SimQFL , a novel simulator that enables decentralized quantum model training across multiple simulated clients by integrating the concepts of quantum machine learning and federated learning for a more practical deployment. The simulator is visually appealing and easy to use for seamless execution. • SimQFL does not require a deep understanding of quantum federated learning; users may upload and train on their own datasets. This feature greatly increases accessibility, allowing researchers with various amounts of experience to conduct custom QFL training.