Deep Probabilistic Programming
This work addresses the need for more efficient and flexible probabilistic programming tools for researchers and practitioners in machine learning, representing a significant advancement rather than an incremental improvement.
The authors tackled the challenge of making probabilistic programming as flexible and efficient as deep learning by proposing Edward, a Turing-complete language that integrates with TensorFlow, resulting in speedups of at least 35x over Stan and 6x over PyMC3 on a benchmark task.
We propose Edward, a Turing-complete probabilistic programming language. Edward defines two compositional representations---random variables and inference. By treating inference as a first class citizen, on a par with modeling, we show that probabilistic programming can be as flexible and computationally efficient as traditional deep learning. For flexibility, Edward makes it easy to fit the same model using a variety of composable inference methods, ranging from point estimation to variational inference to MCMC. In addition, Edward can reuse the modeling representation as part of inference, facilitating the design of rich variational models and generative adversarial networks. For efficiency, Edward is integrated into TensorFlow, providing significant speedups over existing probabilistic systems. For example, we show on a benchmark logistic regression task that Edward is at least 35x faster than Stan and 6x faster than PyMC3. Further, Edward incurs no runtime overhead: it is as fast as handwritten TensorFlow.