PaperlayerView exact revision on arXiv

Source-linked AI summary

Quantum-Assisted Memory-Efficient Training for Parameter-Intensive Wi-Fi-Based Human Activity Recognition

To Truong An, Jie Zhang, Guolin Yin, Junqing Zhang, Yanjiao Li, Trung Q. Duong, Simon L. Cotton

arXiv:2609.04271v1cs.LG

TL;DR

DL-based Wi-Fi HAR is costly in memory during training and inference, limiting deployment on resource-constrained devices. Q-MET uses hybrid quantum-classical parameter generation and in-training structured pruning to address both stages. Across UT-HAR and Widar3.0, it reduces trainable parameters by 90% to 95% and produces 75% to 85% sparse models with limited accuracy loss.

  • Problem

    DL-based HAR requires substantial memory for parameters, optimizer states, activations, and gradients, while existing compression methods mainly target inference after full-scale training.

  • Method

    Q-MET uses a hybrid quantum-classical neural network to generate classical model parameters and applies structured pruning directly during training.

  • Results

    Q-MET reduces trainable parameters by 90% to 95% and achieves 75% to 85% sparsity while preserving or slightly improving classification performance across UT-HAR and Widar3.0.

  • Takeaways & Limitations

    Q-MET jointly improves training and inference efficiency for compact HAR models suited to resource-constrained deployment.

Abstract

from arXiv · show

Wi-Fi-based human activity recognition (HAR) has become an important part of integrated sensing and communications, paving the way for a range of context-aware services. However, most existing Wi-Fi-based HAR systems rely on deep learning (DL) models that are computationally and memory intensive in both training and inference, which poses significant challenges for real-world deployment. Conventional training requires simultaneous updates of millions of parameters, leading to prohibitive memory consumption. In this paper, we propose a novel quantum-assisted memory-efficient training framework (Q-MET) designed to improve efficiency in both training and inference. Q-MET utilizes a hybrid quantum classical neural network to indirectly generate parameters for HAR models, significantly reducing the trainable parameter count compared to direct optimization. To further support the deployment on resource-constrained devices, we integrate structured pruning during the training phase. Experimental results demonstrate that Q-MET achieves a 90% to 95% reduction in trainable parameters compared with conventional backpropagation-based DL training while maintaining or even exceeding classical classification accuracy. Additionally, Q-MET supports lightweight inference through structured pruning, achieving 75% to 85% model sparsity with less than 2% loss in classification accuracy. To the best of our knowledge, this work represents the first quantum-assisted approach to simultaneously tackle memory inefficiencies in both the training and inference stages of HAR systems.

I. Introduction

Wi-Fi sensing supports practical HAR, but DL-based systems remain memory-intensive, especially during training. Q-MET addresses this gap with quantum-assisted parameter generation and pruning integrated into training for compact deployment.

  • Wi-Fi sensing extracts environmental information using existing infrastructure, supporting reliable and scalable HAR applications.
  • DL models learn discriminative Wi-Fi HAR representations but require substantial memory on resource-constrained edge and COTS devices.Training additionally stores parameters, optimizer states, activations, and gradients.
  • Inference-focused compression methods typically train full-scale models before pruning or quantization, leaving training memory costs unresolved.
  • Q-MET uses a hybrid quantum-classical neural network to generate classical DL model parameters with fewer trainable parameters.The framework extends QuantumTrain with a quantum neural network and classical mapping network.
  • Structured pruning is applied during Q-MET training, producing compact models directly deployable for on-device inference without post-training compression.
  • Q-MET achieves up to 95% fewer trainable parameters and 75% to 85% sparsity while preserving or slightly enhancing classification performance.The evaluation uses the Widar3.0 and UT-HAR public datasets.

A. Wi-Fi-based Human Activity Recognition Systems

Wi-Fi HAR commonly uses CSI and DL to capture activity-related signal changes, while quantum-assisted approaches and pruning target efficiency from different directions. Q-MET-related work connects parameter-efficient training with model compression for constrained deployment.

  • Wi-Fi-based Human Activity Recognition Systems: Wi-Fi HAR analyzes wireless-signal changes to obtain contextual information about human activities.
  • Pruning: Prior HAR compression studies use quantization or channel pruning to reduce model size or parameters, primarily for inference efficiency.One reported approach reduced model size by up to 72%, while another reduced parameters by over 50%.
  • Quantum Machine Learning: QML applies quantum computing properties such as superposition and entanglement to machine-learning tasks involving complex data.
  • Quantum Machine Learning: A parameterized quantum circuit uses tunable gate parameters optimized through a hybrid quantum-classical loop.
  • Quantum Machine Learning: A PQC produces a quantum state whose computational-basis measurement yields a probability vector for classical post-processing.For n qubits, measurement provides 2^n possible outcomes.
  • Pruning: Pruning reduces redundant parameters for deployment, but unstructured sparsity can hinder hardware acceleration because weight matrices become irregular.

IV. On-Device Wi-Fi–Based Human Activity Recognition Systems

The on-device Wi-Fi HAR system collects CSI from indoor activities, trains models in the cloud, and deploys compressed models for real-time recognition on resource-constrained devices.

  • System workflow: Wi-Fi-based HAR uses CSI to characterize human movements through channel fluctuations captured between wireless transmitters and receivers.CSI represents subcarrier-level amplitude and phase information derived from the channel frequency response.
  • System workflow: The deployment pipeline comprises dataset preparation, cloud-based training, and on-device inference.CSI is collected indoors, transferred to a cloud server for training, and used by deployed devices for real-time activity classification.
  • System workflow: The model is compressed before deployment to reduce its size for target devices.The described compression step includes pruning before real-time inference on routers or IoT nodes.
  • Signal processing: The two datasets use dataset-specific input preparation before entering a shared ResNet-18 backbone.UT-HAR uses min–max normalization of 1×250×90 samples, whereas Widar3.0 uses standardization with its training-set mean and standard deviation.

B. Deep Learning Model

The study uses ResNet-18 as its representative HAR model and equips Q-MET with a quantum parameter generator that maps quantum outputs into classical network parameters.

  • Deep learning model: ResNet-18 is selected because benchmarking identifies it as highly accurate on both UT-HAR and Widar3.0.The model contains about 11.6 million trainable parameters and uses dataset-specific initial blocks for differing input shapes.
  • Quantum parameter generation: The Q-MET quantum parameter generator combines a parameterized quantum circuit, sinusoidal embedding, and classical mapping network.The mapping network transforms embedded quantum measurement probabilities into parameters for ResNet-18.
  • Quantum parameter generation: The PQC uses U3 gates for qubit-state manipulation and CU3 gates for inter-qubit entanglement.Its circular layout supports multi-qubit interactions, with 6Nq trainable PQC parameters.
  • Quantum parameter generation: Sinusoidal embedding provides compact identifiers for computational-basis states while reducing embedding dimensionality from exponential to constant size.The embedded probabilities are represented through the triplet [S0, S1, Ψ] before mapping into classical parameters.
  • Quantum parameter generation: The mapping network uses inner and outer networks to produce a matrix of classical parameters.The matrix Θm has nG parameter groups with nB parameters per group and is reshaped into the ResNet-18 parameter vector.
  • Quantum parameter generation: Q-MET’s generated parameter vector can be reshaped for ResNet-18 and does not require CSI as input to the quantum generator.This separates parameter generation from sensing modality and supports applicability across classical DL architectures.

B. Structured Pruning

Q-MET applies structured channel pruning using layer-adaptive magnitude-based scores to remove globally low-importance channels while preserving hardware-friendly dense structures.

  • Structured pruning: LAMP-guided channel pruning derives channel importance from the LAMP scores of their associated weights.The method is intended to preserve sensitive layers more effectively than uniform or global magnitude pruning.
  • Structured pruning: Structured pruning is preferred because it produces regular dense matrices suited to hardware acceleration on resource-constrained platforms.Unstructured pruning instead creates irregular sparse matrices that are less practical for deployment.
  • LAMP scoring: LAMP scores measure each weight’s relative importance against higher-magnitude weights within the same layer.The score denominator aggregates squared magnitudes of weights ranked at or above the indexed weight.
  • LAMP scoring: Channel scores are formed by aggregating weight scores, then the lowest-scoring channels are globally removed until the target sparsity is reached.This extends LAMP from individual weights to structured channel removal across the network.

C. Training Flow for Quantum-Assisted Memory-Efficient Training Framework

Q-MET trains a pruned model through warm-up training, structured pruning, and optimal-structure training while dynamically generating parameters for active network components.

  • Training flow: Q-MET training proceeds through warm-up training, structured pruning, and optimal-structure training.These stages establish an initial weight distribution, remove redundant filters, and continue training the compact architecture.
  • Training flow: Warm-up training runs for a few epochs to produce a weight distribution suitable for identifying redundant connections.The warm-up stage is not intended to achieve convergence.
  • Parameter generation: The QPG design selects the PQC qubit count and then configures mapping-network dimensions using the ResNet-18 parameter count.The resulting design specifies nB and nG for parameter generation.
  • Parameter generation: At each iteration, QPG generates parameters only for active unpruned components and injects them into ResNet-18 during the forward computation.PyTorch functional operators accept generated weights and biases for on-the-fly parameter injection.
  • Parameter generation: The generated parameters define the ResNet-18 forward function used to compute the training objective.The training loss for multiclass classification is computed with cross-entropy loss.

2) Loss Function and Gradient Estimation:

Q-MET trains a compact quantum-classical parameter generator instead of directly updating all ResNet-18 parameters, propagating loss gradients through the generator. Its parameter efficiency improves with qubit count, exceeding 90% at six qubits while showing diminishing returns beyond that point.

  • Loss Function and Gradient Estimation: Q-MET propagates the loss gradient through the parameterized quantum circuit and mapping network that generate ResNet-18 parameters.The gradient uses the Jacobian of generated ResNet-18 parameters with respect to the joint generator parameters.
  • Loss Function and Gradient Estimation: Q-MET restricts trainable updates to the compact quantum parameter generator, avoiding direct training of the massive ResNet-18 parameter set.This design reduces training memory overhead in resource-constrained environments.
  • Parameter Efficiency: 90.63% parameter efficiency is achieved with Nq = 6, while Nq = 2 produces a −50% gain and Nq ≥3 reaches approximately 25%.The six-qubit result is derived from the reported relationship between qubit count and parameter efficiency.
  • Parameter Efficiency: 99.41% maximum parameter efficiency is reached at Nq = 10, but increasing Nq from 6 to 10 adds only 8.78% gain.The reported trend indicates diminishing returns after six qubits.

E. Computational Complexity Analysis

Q-MET combines quantum parameter generation, mapping-network computation, and pruned ResNet-18 execution, trading additional computation for substantially lower trainable-state memory. The experiments use two public CSI-based HAR datasets spanning human motions and fine-grained gestures.

  • Complexity Components: Q-MET’s per-iteration complexity combines PQC cost O(2^Nq), mapping-network cost O(nGnB), and pruned-model cost O((1−s)FResNet-18).The sparsity-dependent term scales with the remaining fraction of the full ResNet-18 training cost.
  • Training-Time Trade-off: 18%–36% higher training time is observed because Q-MET adds parameter-generation overhead relative to standard full-model backpropagation.Standard training has complexity O(FResNet-18).
  • Memory Efficiency: Q-MET’s main benefit is memory efficiency: trainable-state memory scales with O(CQ-MET), where CQ-MET ≪ CResNet-18.Standard training stores weights, gradients, and optimizer states for all ResNet-18 parameters.
  • Evaluation Datasets: The evaluation uses UT-HAR and Widar3.0, both containing CSI collected with an Intel 5300 Wi-Fi interface under real-world conditions.UT-HAR covers basic human motions, whereas Widar3.0 targets fine-grained gesture recognition.

B. Training Hyperparameter Configuration

The study defines accuracy, accuracy loss, parameter count, training memory, sparsity, and average epoch time as evaluation measures. Q-MET reduces model-state training memory substantially, while activation and buffer memory remain outside the framework’s primary target.

  • Performance Metrics: Accuracy loss compares Q-MET and baseline ResNet-18 through AccQ-MET and Accbaseline.Accuracy itself is defined as the ratio of correct predictions to total test instances.
  • Efficiency Metrics: Trainable-parameter count is used as the primary efficiency metric for comparing Q-MET with classical training.The paper identifies parameter count as a canonical measure in parameter-efficient training.
  • Training Memory: 95% and 90% model-state memory reductions are reported for UT-HAR and Widar3.0, respectively.Baseline memory is approximately 177 MB versus 8.4 MB for QT-7 on UT-HAR, and 171 MB versus 17 MB for QT-6 on Widar3.0.
  • Memory Scope: Activation and buffer memory is orthogonal to the proposed framework’s target, although structured pruning incidentally reduces it.The reported memory figures quantify parameters, gradients, and optimizer states under FP32.
  • Additional Metrics: Sparsity measures the percentage of the model pruned, while average training time per epoch is measured until convergence.Higher sparsity generally corresponds to reduced model size and faster inference.

3) Sparsity:

The section frames sparsity as a pruning metric and training time as an efficiency measure, while describing experiments across training configurations. The reported setup compares baseline ResNet-18 performance on UT-HAR and Widar3.0.

  • Sparsity: Higher sparsity indicates more aggressive structured pruning and typically corresponds to reduced model size and faster inference.Sparsity is defined as the percentage of the model that has been pruned.
  • Training Time: Average training time per epoch is defined as total training time divided by the number of epochs until convergence.This metric is used to evaluate QT computational efficiency.
  • Experimental Settings: The experiments vary Warm-Up epochs and pruning ratios to analyze Q-MET performance-efficiency trade-offs.The evaluation also examines QT configurations before studying Q-MET settings.
  • Baseline Performance: Table VI reports baseline ResNet-18 performance on UT-HAR and Widar3.0.The table is identified as a baseline-performance summary for the two datasets.

1) Performance of Classical ResNet-18 Model and QT Framework:

QT reduces ResNet-18 trainable parameters substantially while preserving or improving UT-HAR accuracy, though aggressive compression degrades performance on complex Widar3.0 data. The framework also incurs higher training time, while pruning produces high sparsity.

  • Classical baseline: 98.08 ± 0.44% average accuracy was achieved by baseline ResNet-18 on UT-HAR, versus 71.29 ± 1.02% on Widar3.0.Widar3.0 contains more gesture classes and is more complex than UT-HAR.
  • QT performance: 95.29% parameter reduction with QT-7 accompanied 99.08 ± 0.24% average UT-HAR accuracy, exceeding the baseline.QT-5 also reduced parameters by approximately 81.22% while reaching 99.20 ± 0.22%.
  • Training efficiency: 35.7% higher training time at QT-10 than the baseline accompanied the parameter-efficiency gains.The baseline required 3.2 seconds per epoch, while QT-3 already incurred an 18.2% overhead.
  • QT performance: 90%+ parameter reduction with QT-6 preserved 70.61 ± 0.45% accuracy on Widar3.0, whereas QT-9 and QT-10 fell to 66.28 ± 1.94% and 65.84 ± 2.03%.Increasing variance at higher qubit counts indicates reduced stability and possible underfitting.
  • Controlled comparisons: QT-7 outperformed a parameter-matched lightweight ResNet-18 by 1.48 percentage points on UT-HAR and 6.16 points on Widar3.0.The comparison used a slim ResNet-18 retaining 22% of each convolutional layer’s channels.
  • Pruning: 90% sparsity was reached at a pruning ratio of 0.9, with sparsity exceeding 70% above a ratio of 0.6.Sparsity increased monotonically with the pruning ratio on both datasets.

5) Q-MET Performance:

Q-MET integrates structured pruning into quantum-assisted training to improve deployment efficiency while retaining competitive recognition accuracy. Across datasets, it achieves high sparsity with modest accuracy losses, although Widar3.0 is more sensitive to aggressive pruning.

  • Pruning performance: 98.44%+ UT-HAR accuracy was maintained at a 0.6 pruning ratio while more than 70% of parameters were removed.Accuracy remained above 98.4% across pruning ratios from 0.1 to 0.8.
  • Pruning performance: 70.02% peak Q-MET accuracy on Widar3.0 at a 0.2 pruning ratio was close to the QT-6 baseline’s 70.61%.Accuracy declined to 66.01% at a pruning ratio of 0.9, with higher variance under extreme sparsity.
  • Train-then-prune comparison: 50 extra fine-tuning epochs were required for TTP50 to approach Q-MET at a 0.7 pruning ratio, while Q-MET needed no post-pruning training.At that ratio, TTP50 reached 98.80% on UT-HAR and 64.74% on Widar3.0, compared with Q-MET’s 98.84% and 66.72%.
  • Inference efficiency: Q-MET reached 85%+ UT-HAR sparsity at a 0.8 pruning ratio with 98.60% accuracy, within 1% of unpruned QT-7.The pruned model still exceeded the classical unpruned baseline’s 98.08%.
  • Inference efficiency: 75% Widar3.0 sparsity at a 0.6 pruning ratio yielded 69.20% accuracy, 1.42 points below unpruned QT-6.A single warm-up epoch was sufficient for steady pruning performance.
  • Overall results: 95% fewer trainable parameters on UT-HAR and 90% fewer on Widar3.0 were reported alongside approximately 85% and 75% model sparsity, respectively.Accuracy improved from 98.08% to 99.08% on UT-HAR and decreased from 71.29% to 70.61% on Widar3.0.
Loading 2609.04271v1…