X Liu, C Gong, Q Liu - arXiv preprint arXiv:2209.03003, 2022 - arxiv.org
We present rectified flow, a surprisingly simple approach to learning (neural) ordinary differential equation (ODE) models to transport between two empirically observed …
Recently, Optimal Transport has been proposed as a probabilistic framework in Machine Learning for comparing and manipulating probability distributions. This is rooted in its rich …
Optimal transport has recently been reintroduced to the machine learning community thanks in part to novel efficient optimization procedures allowing for medium to large scale …
Simulation-free methods for training continuous-time generative models construct probability paths that go between noise distributions and individual data samples. Recent works, such …
In computer vision, one is often confronted with problems of domain shifts, which occur when one applies a classifier trained on a source dataset to target data sharing similar …
H Wang, J Fan, Z Chen, H Li, W Liu… - Advances in …, 2024 - proceedings.neurips.cc
Estimating individual treatment effects from observational data is challenging due to treatment selection bias. Prevalent methods mainly mitigate this issue by aligning different …
M Li, YM Zhai, YW Luo, PF Ge… - Proceedings of the …, 2020 - openaccess.thecvf.com
Unsupervised domain adaptation (UDA) is a representative problem in transfer learning, which aims to improve the classification performance on an unlabeled target domain by …
In this paper, we present a novel and principled approach to learn the optimal transport between two distributions, from samples. Guided by the optimal transport theory, we learn …
We present a novel neural-networks-based algorithm to compute optimal transport maps and plans for strong and weak transport costs. To justify the usage of neural networks, we …