Pith. sign in

REVIEW 8 cited by

AirPhyNet: Harnessing Physics-Guided Neural Networks for Air Quality Prediction

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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

arxiv 2402.03784 v2 pith:2WC2UAEG submitted 2024-02-06 cs.LG cs.AIphysics.app-ph

Kethmi Hirushini Hettige , Jiahao Ji , Shili Xiang , Cheng Long , Gao Cong , Jingyuan Wang This is my paper · ORCID
classification cs.LGcs.AIphysics.app-ph
keywords predictionqualityairphynetdataneuralphysicalphysicsdifferent
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
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.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Towards Propagation-aware Representation Learning for Supervised Social Media Graph Analytics

    cs.SI 2025-09 conditional novelty 6.0 of 10

    A single graph-representation framework with a kinetic-model loss beats task-specific baselines on rumor, bot, and cascade tasks and transfers across datasets.

  2. GTG: Generalizable Trajectory Generation Model for Urban Mobility

    cs.LG 2025-02 reject novelty 6.0 of 10

    GTG combines graph networks, adversarial domain adaptation, and preference-based shortest path search to generate trajectories in cities without historical trajectory data, but its evaluation leaks the test origins an...

  3. BIGCity: A Universal Spatiotemporal Model for Unified Trajectory and Traffic State Data Analysis

    cs.AI 2024-12 conditional novelty 6.0 of 10

    One model, BIGCity, represents both individual trajectories and population-level traffic states as sequences of shared ST-units and performs eight spatiotemporal tasks with a single set of weights, reportedly beating ...

  4. A Modular Multitask Reasoning Framework Integrating Spatio-temporal Models and LLMs

    cs.CL 2025-06 conditional novelty 5.0 of 10

    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.

  5. Air in Your Neighborhood: Fine-Grained AQI Forecasting Using Mobile Sensor Data

    cs.LG 2025-06 conditional novelty 5.0 of 10

    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.

  6. Graph-Based Physics-Guided Urban PM2.5 Air Quality Imputation with Constrained Monitoring Data

    cs.LG 2025-06 conditional novelty 5.0 of 10

    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.

  7. Epidemiology-informed Network for Robust Rumor Detection

    cs.SI 2024-11 conditional novelty 4.0 of 10

    EIN augments graph rumor detectors with an eUSD-style state encoder trained against LLM-generated stance labels, achieving 0.6 to 1.6 percentage point accuracy gains over RAGCL and better depth-robustness on DRWeibo, ...

  8. Mass-Conserving Physics-Informed Neural Networks For The One-Dimensional Advection-Diffusion Equation

    physics.comp-ph 2026-07 conditional novelty 3.0 of 10

    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.

Pith tools