The Exponentially Weighted Signature

• URL: http://arxiv.org/abs/2603.19198v1
• Date: Thu, 19 Mar 2026 17:51:20 GMT
• Title: The Exponentially Weighted Signature
• Abstract summary: We introduce the Exponentially Weighted Signature (EWS)<n>We show that the EWS is the unique solution to a linear controlled differential equation on the tensor algebra.<n>We use this framework to empirically demonstrate the expressivity gap between the EWS and both the signature and EFM on two SDE-based regression tasks.
• Score: 41.02458927230264
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The signature is a canonical representation of a multidimensional path over an interval. However, it treats all historical information uniformly, offering no intrinsic mechanism for contextualising the relevance of the past. To address this, we introduce the Exponentially Weighted Signature (EWS), generalising the Exponentially Fading Memory (EFM) signature from diagonal to general bounded linear operators. These operators enable cross-channel coupling at the level of temporal weighting together with richer memory dynamics including oscillatory, growth, and regime-dependent behaviour, while preserving the algebraic strengths of the classical signature. We show that the EWS is the unique solution to a linear controlled differential equation on the tensor algebra, and that it generalises both state-space models and the Laplace and Fourier transforms of the path. The group-like structure of the EWS enables efficient computation and makes the framework amenable to gradient-based learning, with the full semigroup action parametrised by and learned through its generator. We use this framework to empirically demonstrate the expressivity gap between the EWS and both the signature and EFM on two SDE-based regression tasks.

Related papers

• Adaptive Anisotropic Attention for Axis-Structured Signals [5.616707402426108]
  Self-attention treats all token pairs as equally plausible before learning, an interaction-isotropic prior that can be mismatched to structured signals.<n>We introduce Adaptive Anisotropic Attention (AAA), which splits attention into two paths: a temporal path and a spatial path.<n>We show that both paths are necessary and that the weighted sum beats a hard choice of one path; most of the benefit comes from the gate learning a different temporal/spatial balance at each layer of the network.
  arXiv  Detail & Related papers  (2026-09-08T14:19:33Z)
• Accelerating Unified Multimodal Models with Core-Expansion Routing and Unified Computation Scheduling [61.28599393707943]
  Unified multimodal models jointly support understanding and generation, but incur substantial redundant computation across tokens, layers, and generation timesteps.<n>We propose a task-shared core scorer and progress-conditioned generation expansions, optimized through generation decomposition and cross-task core alignment.<n> Experiments on two representative UMM architectures demonstrate consistent quality--efficiency improvements across both tasks, retaining 98.03% of dense understanding performance with a 1.93times$ end-to-end inference speedup.
  arXiv  Detail & Related papers  (2026-08-29T14:27:55Z)
• Generative Path-Law Jump-Diffusion: Sequential MMD-Gradient Flows and Generalisation Bounds in Marcus-Signature RKHS [0.0]
  This paper introduces a novel generative framework for synthesising cdlg trajectories that are sequentially consistent with time-evolving path-law proxies.<n>We develop the textitAnticipatory Neural Jump-Diffusion (ANJD) flow, a generative mechanism that effectively inverts the time-extended Marcus-sense signature.
  arXiv  Detail & Related papers  (2026-04-06T13:23:59Z)
• Rethinking Diffusion Models with Symmetries through Canonicalization with Applications to Molecular Graph Generation [56.361076943802594]
  CanonFlow achieves state-of-the-art performance on the challenging GEOM-DRUG dataset, and the advantage remains large in few-step generation.
  arXiv  Detail & Related papers  (2026-02-16T18:58:55Z)
• From Path Signatures to Sequential Modeling: Incremental Signature Contributions for Offline RL [1.278093617645299]
  We propose the Incremental Signature Contribution (ISC) method, which decomposes truncated path signatures into a temporally ordered sequence of elements.<n>ISC is inherently sensitive to instantaneous trajectory updates, which is critical for sensitive and stability-requiring control dynamics.<n>We introduce ISC-Transformer (ISCT), an offline reinforcement learning model that integrates ISC into a standard Transformer architecture.
  arXiv  Detail & Related papers  (2026-02-12T10:37:37Z)
• SigMA: Path Signatures and Multi-head Attention for Learning Parameters in fBm-driven SDEs [0.0]
  SigMA learns model parameters from synthetically generated paths of fBm-driven differential equations.<n>It consistently outperforms CNN, LSTM, vanilla Transformer, and Deep Signature baselines in accuracy, robustness, and model compactness.
  arXiv  Detail & Related papers  (2025-12-17T05:09:18Z)
• Beyond Log-Concavity and Score Regularity: Improved Convergence Bounds for Score-Based Generative Models in W2-distance [0.0]
  We present a novel framework for analyzing $mathcalW$-convergence in Generative Models (SGMs)<n>We show that weak log-concavity of the data distribution evolves into log-concavity over time.<n>Our approach circumvents the need for stringent regularity conditions on the score function and its estimators.
  arXiv  Detail & Related papers  (2025-01-04T14:33:27Z)
• Degree-Conscious Spiking Graph for Cross-Domain Adaptation [51.58506501415558]
  Spiking Graph Networks (SGNs) have demonstrated significant potential in graph classification.<n>We introduce a novel framework named Degree-Consicious Spiking Graph for Cross-Domain Adaptation (DeSGraDA)<n>DeSGraDA enhances generalization across domains with three key components.
  arXiv  Detail & Related papers  (2024-10-09T13:45:54Z)
• LEAF: Unveiling Two Sides of the Same Coin in Semi-supervised Facial Expression Recognition [56.22672276092373]
  Semi-supervised learning has emerged as a promising approach to tackle the challenge of label scarcity in facial expression recognition.<n>We propose a unified framework termed hierarchicaL dEcoupling And Fusing (LEAF) to coordinate expression-relevant representations and pseudo-labels for semi-supervised FER.
  arXiv  Detail & Related papers  (2024-04-23T13:43:33Z)
• SNEkhorn: Dimension Reduction with Symmetric Entropic Affinities [14.919246099820548]
  Entropic affinities (EAs) are used in the popular Dimensionality Reduction (DR) algorithm t-SNE. EAs are inherently asymmetric and row-wise, but they are used in DR approaches after undergoing symmetrization methods. In this work, we uncover a novel characterization of EA as an optimal transport problem, allowing a natural symmetrization that can be computed efficiently.
  arXiv  Detail & Related papers  (2023-05-23T08:08:10Z)
• Adaptive Fourier Neural Operators: Efficient Token Mixers for Transformers [55.90468016961356]
  We propose an efficient token mixer that learns to mix in the Fourier domain. AFNO is based on a principled foundation of operator learning. It can handle a sequence size of 65k and outperforms other efficient self-attention mechanisms.
  arXiv  Detail & Related papers  (2021-11-24T05:44:31Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.