Structured Output Learning with Conditional Generative Flows
This work addresses the challenge of training structured prediction models for researchers and practitioners by providing a versatile method that avoids approximations, though it is incremental as it builds on existing flow-based models.
The paper tackles the problem of intractable likelihood in structured prediction by proposing conditional Glow (c-Glow), a conditional generative flow model that computes p(y|x) exactly and efficiently, eliminating the need for surrogate objectives or inference during training. It outperforms state-of-the-art baselines in some tasks and achieves comparable results in others across five different structured prediction tasks.
Traditional structured prediction models try to learn the conditional likelihood, i.e., p(y|x), to capture the relationship between the structured output y and the input features x. For many models, computing the likelihood is intractable. These models are therefore hard to train, requiring the use of surrogate objectives or variational inference to approximate likelihood. In this paper, we propose conditional Glow (c-Glow), a conditional generative flow for structured output learning. C-Glow benefits from the ability of flow-based models to compute p(y|x) exactly and efficiently. Learning with c-Glow does not require a surrogate objective or performing inference during training. Once trained, we can directly and efficiently generate conditional samples. We develop a sample-based prediction method, which can use this advantage to do efficient and effective inference. In our experiments, we test c-Glow on five different tasks. C-Glow outperforms the state-of-the-art baselines in some tasks and predicts comparable outputs in the other tasks. The results show that c-Glow is versatile and is applicable to many different structured prediction problems.