paper-with-me

홈 › Papers

Feature-aligned N-BEATS with Sinkhorn divergence

2023-05-24 · Joonhun Lee, Myeongho Jeon, Myungjoo Kang, Kyunghyun Park

We propose Feature-aligned N-BEATS as a domain-generalized time series forecasting model. It is a nontrivial extension of N-BEATS with doubly residual stacking principle (Oreshkin et al. [45]) into a representation learning framework. In particular, it revolves around marginal feature probability measures induced by the intricate composition of residual and feature extracting operators of N-BEATS in each stack and aligns them stack-wise via an approximate of an optimal transport distance referred to as the Sinkhorn divergence. The training loss consists of an empirical risk minimization from multiple source domains, i.e., forecasting loss, and an alignment loss calculated with the Sinkhorn divergence, which allows the model to learn invariant features stack-wise across multiple source data sequences while retaining N-BEATS's interpretable design and forecasting power. Comprehensive experimental evaluations with ablation studies are provided and the corresponding results demonstrate the proposed model's forecasting and generalization capabilities.

📄 PDF Abstract BibTeX arXiv:2305.15196

Code (1)

leejoonhun/fan-beats 공식 구현 pytorch

Tasks

Domain GeneralizationRepresentation LearningTime SeriesTime Series ForecastingUnivariate Time Series Forecasting

Similar Papers 제목 키워드 기반

Hilbert Sinkhorn Divergence for Optimal Transport

2021-06-19 · CVPR 2021 1 · Qian Li, Zhichao Wang, Gang Li, Jun Pang 외

Sinkhorn divergence has become a very popular metric to compare probability distributions in optimal transport. However, most works resort to Sinkhorn divergence in Euclidean space, which greatly blocks their applica…

image-classificationImage ClassificationTopological Data Analysis

Distributional Reinforcement Learning with Regularized Wasserstein Loss

2022-02-01 · Ke Sun, Yingnan Zhao, Wulong Liu, Bei Jiang 외

The empirical success of distributional reinforcement learning (RL) highly relies on the choice of distribution divergence equipped with an appropriate distribution representation. In this paper, we propose \textit{Sinkh…

Atari GamesDistributional Reinforcement Learningreinforcement-learningReinforcement Learning+1

Convergence and finite sample approximations of entropic regularized Wasserstein distances in Gaussian and RKHS settings

2021-01-05 · Minh Ha Quang

This work studies the convergence and finite sample approximations of entropic regularized Wasserstein distances in the Hilbert space setting. Our first main result is that for Gaussian measures on an infinite-dimensiona…

Sinkhorn-Drifting Generative Models

2026-03-12 · Ping He, Om Khangaonkar, Hamed Pirsiavash, Yikun Bai 외 arxiv

We establish a theoretical link between the recently proposed "drifting" generative dynamics and gradient flows induced by the Sinkhorn divergence. In a particle discretization, the drift field admits a cross-minus-self …

Optimal transport with $f$-divergence regularization and generalized Sinkhorn algorithm

2021-05-29 · Dávid Terjék, Diego González-Sánchez

Entropic regularization provides a generalization of the original optimal transport problem. It introduces a penalty term defined by the Kullback-Leibler divergence, making the problem more tractable via the celebrated S…