REVIEW 5 cited by
AirPhyNet: Harnessing Physics-Guided Neural Networks for Air Quality Prediction
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Air quality prediction and modelling plays a pivotal role in public health and environment management, for individuals and authorities to make informed decisions. Although traditional data-driven models have shown promise in this domain, their long-term prediction accuracy can be limited, especially in scenarios with sparse or incomplete data and they often rely on black-box deep learning structures that lack solid physical foundation leading to reduced transparency and interpretability in predictions. To address these limitations, this paper presents a novel approach named Physics guided Neural Network for Air Quality Prediction (AirPhyNet). Specifically, we leverage two well-established physics principles of air particle movement (diffusion and advection) by representing them as differential equation networks. Then, we utilize a graph structure to integrate physics knowledge into a neural network architecture and exploit latent representations to capture spatio-temporal relationships within the air quality data. Experiments on two real-world benchmark datasets demonstrate that AirPhyNet outperforms state-of-the-art models for different testing scenarios including different lead time (24h, 48h, 72h), sparse data and sudden change prediction, achieving reduction in prediction errors up to 10%. Moreover, a case study further validates that our model captures underlying physical processes of particle movement and generates accurate predictions with real physical meaning.
Forward citations
Cited by 5 Pith papers
-
Towards Propagation-aware Representation Learning for Supervised Social Media Graph Analytics
A single graph-representation framework with a kinetic-model loss beats task-specific baselines on rumor, bot, and cascade tasks and transfers across datasets.
-
A Modular Multitask Reasoning Framework Integrating Spatio-temporal Models and LLMs
STReason uses in-context learning to convert spatio-temporal queries into executable programs with specialized modules, outperforming plain LLMs on a new 150-query benchmark.
-
Air in Your Neighborhood: Fine-Grained AQI Forecasting Using Mobile Sensor Data
A deep learning pipeline on mobile-sensor data achieves much lower AQI forecast error than classical baselines on Delhi's AirDelhi dataset, though the headline 79% improvement is overstated due to metric confusion.
-
Graph-Based Physics-Guided Urban PM2.5 Air Quality Imputation with Constrained Monitoring Data
GraPhy, a physics-inspired graph neural network with wind-based edge features and learnable diffusion scaling, reports the best PM2.5 imputation accuracy among six baselines on 41 sensors in Fresno, California.
-
Mass-Conserving Physics-Informed Neural Networks For The One-Dimensional Advection-Diffusion Equation
Adding a soft mass-conservation penalty to PINNs for the 1D advection-diffusion equation reduces long-term relative L2 error by 9–67× and mass error by 15–215× compared to vanilla PINNs across Peclet numbers 0.01–20.
Discussion (0). Sign in to comment.