Flow Matching Neural Processes
This work provides a more efficient and simpler method for inference and sampling in neural processes, benefiting researchers and practitioners in machine learning, though it is incremental as it builds on existing NP frameworks.
The paper tackled the problem of improving neural processes for learning stochastic processes from data by introducing a flow matching-based model, which outperformed previous state-of-the-art methods on benchmarks including synthetic 1D Gaussian processes, 2D images, and real-world weather data.
Neural processes (NPs) are a class of models that learn stochastic processes directly from data and can be used for inference, sampling and conditional sampling. We introduce a new NP model based on flow matching, a generative modeling paradigm that has demonstrated strong performance on various data modalities. Following the NP training framework, the model provides amortized predictions of conditional distributions over any arbitrary points in the data. Compared to previous NP models, our model is simple to implement and can be used to sample from conditional distributions using an ODE solver, without requiring auxiliary conditioning methods. In addition, the model provides a controllable tradeoff between accuracy and running time via the number of steps in the ODE solver. We show that our model outperforms previous state-of-the-art neural process methods on various benchmarks including synthetic 1D Gaussian processes data, 2D images, and real-world weather data.