Selected work

Representative Papers

Scalable Uncertainty Quantification for Extreme Weather Forecasting via Empirical Neural Tangent Kernels

Jun 01, 2026

This work addresses the critical limitation of existing deep learning weather forecasting models—their lack of reliable uncertainty quantification, which hinders high-stakes decision-making during extreme weather events. Leveraging the neural tangent kernel (NTK) framework, the authors introduce a Gaussian process correction term constructed from empirical features of the final network layer, enabling inference-time uncertainty estimation without model retraining. By uncovering an architecture-dependent variance collapse mechanism, they propose a data-driven decomposition strategy based on the spectral concentration of features, which for the first time yields prediction intervals that adapt to the severity of extreme events. At 90% coverage, the resulting intervals achieve 31–37% improved sharpness over split conformal prediction.

0 citationsRead paper

eXplaining to Learn (eX2L): Regularization Using Contrastive Visual Explanation Pairs for Distribution Shifts

May 07, 2026

Existing methods often exhibit unstable performance under distribution shifts, struggle to surpass empirical risk minimization baselines, and lack interpretability. To address these limitations, this work proposes an explanation-based regularization framework that explicitly decouples label-relevant features from confounding attributes at the group level. The approach constructs explanation pairs by comparing Grad-CAM activation maps generated by a main classifier and a confounder classifier, then introduces a similarity penalty to enforce disentanglement between class-discriminative and spurious features, thereby promoting functional domain invariance. Evaluated on the Spawrious Many-to-Many Hard Challenge, the method achieves an average accuracy of 82.24% and a worst-group accuracy of 66.31%, outperforming the current state of the art by 5.49% and 10.90%, respectively, while simultaneously enhancing both model performance and interpretability.

0 citationsRead paper

Influence Malleability in Linearized Attention: Dual Implications of Non-Convergent NTK Dynamics

Mar 13, 2026

This work uncovers the fundamental reason why linearized attention mechanisms fail to converge within the Neural Tangent Kernel (NTK) framework and elucidates their dual impact on model performance and robustness. By constructing a linearized attention operator that exactly corresponds to the data-dependent Gram kernel, and integrating NTK theory with spectral analysis, the study establishes—for the first time—a theoretical link between the cubic amplification of the Gram matrix condition number and the required network width (m = Ω(κ⁶)). It introduces the notion of “influence plasticity” to unify the explanation of attention’s expressive power and its vulnerability to adversarial perturbations. Empirical results show that practical training widths fall far below the theoretical threshold, yet exhibit 6–9 times higher influence plasticity than ReLU networks, conferring superior task adaptability at the cost of heightened sensitivity to data poisoning.

0 citationsRead paper

UltraLIF: Fully Differentiable Spiking Neural Networks via Ultradiscretization and Max-Plus Algebra

Feb 10, 2026

This work addresses the challenge of training spiking neural networks (SNNs), which typically rely on heuristic surrogate gradients due to the non-differentiability of spikes, leading to training instability and forward–backward inconsistency. For the first time, the authors introduce hyper-discretization from tropical geometry into SNNs, constructing a differentiable continuous relaxation model via max-plus algebra and a log-sum-exp soft thresholding operator. They rigorously establish its theoretical convergence to the Leaky Integrate-and-Fire (LIF) neuron dynamics, thereby eliminating the need for surrogate gradients and enabling standard backpropagation with spatiotemporal modeling. Experiments demonstrate consistent superiority over existing surrogate gradient methods across six benchmarks—including static images, neuromorphic vision, and audio tasks—with notable performance gains under single-timestep (T=1) settings and significant energy savings achievable through sparsity-inducing penalties.

0 citationsRead paper

SPIKE: Sparse Koopman Regularization for Physics-Informed Neural Networks

Jan 15, 2026

This work addresses the poor spatiotemporal extrapolation and generalization of physics-informed neural networks (PINNs), which often suffer from overfitting within the training domain. To mitigate this, the authors propose the SPIKE framework, which regularizes PINNs by incorporating a continuous-time Koopman operator to enforce linear dynamics in a learned observable space, yielding a compact and structured representation of the underlying system. By integrating L1 sparse regularization, SPIKE learns a sparse generator matrix that reflects the intrinsic low-dimensional simplicity of complex dynamical systems while avoiding the diagonal dominance commonly observed in discrete Koopman operators. Experiments on diverse partial differential equations—including Navier–Stokes—and chaotic ordinary differential equations such as the Lorenz system demonstrate that SPIKE substantially improves temporal extrapolation, spatial generalization, long-term prediction accuracy, and achieves unconditional stability.

0 citationsRead paper

Recent publications

Latest Papers

Scalable Uncertainty Quantification for Extreme Weather Forecasting via Empirical Neural Tangent Kernels

Jun 01, 2026

This work addresses the critical limitation of existing deep learning weather forecasting models—their lack of reliable uncertainty quantification, which hinders high-stakes decision-making during extreme weather events. Leveraging the neural tangent kernel (NTK) framework, the authors introduce a Gaussian process correction term constructed from empirical features of the final network layer, enabling inference-time uncertainty estimation without model retraining. By uncovering an architecture-dependent variance collapse mechanism, they propose a data-driven decomposition strategy based on the spectral concentration of features, which for the first time yields prediction intervals that adapt to the severity of extreme events. At 90% coverage, the resulting intervals achieve 31–37% improved sharpness over split conformal prediction.

0 citationsRead paper

eXplaining to Learn (eX2L): Regularization Using Contrastive Visual Explanation Pairs for Distribution Shifts

May 07, 2026

Existing methods often exhibit unstable performance under distribution shifts, struggle to surpass empirical risk minimization baselines, and lack interpretability. To address these limitations, this work proposes an explanation-based regularization framework that explicitly decouples label-relevant features from confounding attributes at the group level. The approach constructs explanation pairs by comparing Grad-CAM activation maps generated by a main classifier and a confounder classifier, then introduces a similarity penalty to enforce disentanglement between class-discriminative and spurious features, thereby promoting functional domain invariance. Evaluated on the Spawrious Many-to-Many Hard Challenge, the method achieves an average accuracy of 82.24% and a worst-group accuracy of 66.31%, outperforming the current state of the art by 5.49% and 10.90%, respectively, while simultaneously enhancing both model performance and interpretability.

0 citationsRead paper

Influence Malleability in Linearized Attention: Dual Implications of Non-Convergent NTK Dynamics

Mar 13, 2026

This work uncovers the fundamental reason why linearized attention mechanisms fail to converge within the Neural Tangent Kernel (NTK) framework and elucidates their dual impact on model performance and robustness. By constructing a linearized attention operator that exactly corresponds to the data-dependent Gram kernel, and integrating NTK theory with spectral analysis, the study establishes—for the first time—a theoretical link between the cubic amplification of the Gram matrix condition number and the required network width (m = Ω(κ⁶)). It introduces the notion of “influence plasticity” to unify the explanation of attention’s expressive power and its vulnerability to adversarial perturbations. Empirical results show that practical training widths fall far below the theoretical threshold, yet exhibit 6–9 times higher influence plasticity than ReLU networks, conferring superior task adaptability at the cost of heightened sensitivity to data poisoning.

0 citationsRead paper

UltraLIF: Fully Differentiable Spiking Neural Networks via Ultradiscretization and Max-Plus Algebra

Feb 10, 2026

This work addresses the challenge of training spiking neural networks (SNNs), which typically rely on heuristic surrogate gradients due to the non-differentiability of spikes, leading to training instability and forward–backward inconsistency. For the first time, the authors introduce hyper-discretization from tropical geometry into SNNs, constructing a differentiable continuous relaxation model via max-plus algebra and a log-sum-exp soft thresholding operator. They rigorously establish its theoretical convergence to the Leaky Integrate-and-Fire (LIF) neuron dynamics, thereby eliminating the need for surrogate gradients and enabling standard backpropagation with spatiotemporal modeling. Experiments demonstrate consistent superiority over existing surrogate gradient methods across six benchmarks—including static images, neuromorphic vision, and audio tasks—with notable performance gains under single-timestep (T=1) settings and significant energy savings achievable through sparsity-inducing penalties.

0 citationsRead paper

SPIKE: Sparse Koopman Regularization for Physics-Informed Neural Networks

Jan 15, 2026

This work addresses the poor spatiotemporal extrapolation and generalization of physics-informed neural networks (PINNs), which often suffer from overfitting within the training domain. To mitigate this, the authors propose the SPIKE framework, which regularizes PINNs by incorporating a continuous-time Koopman operator to enforce linear dynamics in a learned observable space, yielding a compact and structured representation of the underlying system. By integrating L1 sparse regularization, SPIKE learns a sparse generator matrix that reflects the intrinsic low-dimensional simplicity of complex dynamical systems while avoiding the diagonal dominance commonly observed in discrete Koopman operators. Experiments on diverse partial differential equations—including Navier–Stokes—and chaotic ordinary differential equations such as the Lorenz system demonstrate that SPIKE substantially improves temporal extrapolation, spatial generalization, long-term prediction accuracy, and achieves unconditional stability.

0 citationsRead paper