10 April 2025 in Pattern Recognition
12 February 2026 in Journal of the Royal Statistical Society Series A: Statistics in Society
Modelling and forecasting air quality is crucial for effective air pollution management and protecting public health. Air quality data, characterized by nonlinearity, nonstationarity, and spatio-temporal correlations, often include extreme pollutant levels in severely polluted cities (e.g. Delhi, the capital of India). This is ignored by various geometric deep learning models, such as spatio-temporal graph convolutional networks (STGCNs), which are otherwise effective for spatio-temporal forecasting. This study develops an extreme value theory (EVT) guided modified STGCN model (E-STGCN) for air pollution data to incorporate extreme behaviour across pollutant concentrations. E-STGCN combines graph convolutional networks for spatial modelling and EVT-guided long short-term memory units for temporal sequence learning. Along with spatial and temporal components, it incorporates a generalized Pareto distribution to capture the extreme behaviour of different air pollutants and embed this information into the learning process. The proposal is then applied to analyse air pollution data of 37 monitoring stations across Delhi, India. The forecasting performance for different test horizons is compared to benchmark forecasters (both temporal and spatio-temporal). It is found that E-STGCN has consistent performance across all seasons. The robustness of our results has also been evaluated empirically. Moreover, combined with conformal prediction, E-STGCN can produce probabilistic prediction intervals.
12 December 2024 in IEEE Access
Spatiotemporal forecasting of traffic flow data represents a typical problem impacting urban traffic management systems. In general, spatiotemporal forecasting problems involve complex interactions, nonlinearities, and long-range dependencies due to the interwoven nature of the temporal and spatial dimensions. Due to this, traditional statistical and machine learning methods cannot adequately handle both the temporal and spatial dependencies in traffic flow datasets. A prevalent approach in the field combines graph convolutional networks and multi-head attention mechanisms for spatiotemporal processing of traffic flow data. This paper proposes a wavelet-based temporal attention model, namely a wavelet-based dynamic spatiotemporal aware graph neural networks (W-DSTAGNN), for tackling the traffic forecasting problem. A key feature of W-DSTAGNN is the use of wavelets that can effectively separate noise and capture multi-resolution temporal patterns. Wavelet decomposition can help by decomposing the signal into components that can be analyzed independently, reducing the impact of non-stationarity and handling long-range dependencies of traffic flow datasets. These enable our proposal to generate more robust and better accurate forecasts of complex and dynamic traffic dependencies than commonly used spatiotemporal deep learning models. Benchmark experiments using three popularly used statistical metrics confirm that our proposal efficiently captures spatiotemporal correlations and outperforms eleven state-of-the-art models (including both temporal and spatiotemporal benchmarks) on three publicly available traffic datasets. Our proposed ensemble method can better handle dynamic temporal and spatial dependencies and make reliable long-term forecasts. In addition to point forecasts, our proposed model can generate interval forecasts that significantly enhance probabilistic forecasting of traffic datasets.
1 November 2025 in Chaos: An Interdisciplinary Journal of Nonlinear Science
Understanding and forecasting precipitation events in the Arctic maritime environments, such as Bear Island and Ny-Ålesund, is crucial for assessing climate risk and developing early warning systems in vulnerable marine regions. This study proposes a probabilistic machine learning framework for modeling and predicting the dynamics and severity of precipitation. We begin by analyzing the scale-dependent relationships between precipitation and key atmospheric drivers (e.g., temperature, relative humidity, cloud cover, and air pressure) using wavelet coherence, which captures localized dependencies across time and frequency domains. To assess joint causal influences, we employ synergistic-unique-redundant decomposition, which quantifies the impact of interaction effects among each variable on future precipitation dynamics. These insights inform the development of data-driven forecasting models that incorporate both historical precipitation and causal climate drivers. To account for uncertainty, we employ the conformal prediction method, which enables the generation of calibrated non-parametric prediction intervals. Our results underscore the importance of utilizing a comprehensive framework that combines causal analysis with probabilistic forecasting to enhance the reliability and interpretability of precipitation predictions in Arctic marine environments.
29 October 2026 in APL Machine Learning
Chaotic time series are notoriously difficult to forecast: small uncertainties in initial conditions amplify rapidly, while strong nonlinearities and regime-dependent variability constrain predictability. Although modern deep learning often delivers strong short-horizon accuracy, its black-box nature limits scientific insight and practical trust in settings where understanding the underlying dynamics matters. To address this gap, we propose two complementary symbolic forecasters that learn explicit, interpretable algebraic equations from chaotic time series data. Symbolic neural forecaster adapts a neural network-based equation learning architecture to the forecasting setting, enabling end-to-end differentiable discovery of compact and interpretable algebraic relations. The symbolic tree forecaster builds on evolutionary symbolic regression to search directly over equation structures under a principled accuracy–complexity trade-off. We evaluate both approaches in a rolling-window one-step-ahead forecast setting using several accuracy metrics and compare against a broad suite of baselines spanning classical statistical models, tree ensembles, and state-of-the-art deep learning architectures. Numerical experiments cover a benchmark of 132 low-dimensional chaotic attractors and two real-world chaotic time series (weekly dengue incidence in San Juan and the Niño 3.4 sea-surface temperature index). Across datasets, symbolic forecasters achieve competitive one-step-ahead accuracy while providing transparent analytical equations that reveal salient aspects of the underlying dynamics.