Large-Scale Optimal Transport via Adversarial Training with Cycle-Consistency
This addresses the scalability and flexibility limitations of existing optimal transport methods for machine learning practitioners working on tasks requiring distribution alignment.
The paper tackles the problem of large-scale optimal transport by proposing an end-to-end approach that directly solves the transport map and supports general cost functions, using adversarial training with cycle-consistency constraints. It demonstrates superior performance in applications like domain adaptation, image-to-image translation, and color transfer.
Recent advances in large-scale optimal transport have greatly extended its application scenarios in machine learning. However, existing methods either not explicitly learn the transport map or do not support general cost function. In this paper, we propose an end-to-end approach for large-scale optimal transport, which directly solves the transport map and is compatible with general cost function. It models the transport map via stochastic neural networks and enforces the constraint on the marginal distributions via adversarial training. The proposed framework can be further extended towards learning Monge map or optimal bijection via adopting cycle-consistency constraint(s). We verify the effectiveness of the proposed method and demonstrate its superior performance against existing methods with large-scale real-world applications, including domain adaptation, image-to-image translation, and color transfer.