Unifying Generative Models with GFlowNets and Beyond
This work addresses the problem of disparate generative modeling approaches for researchers and practitioners, offering a unifying perspective that is incremental in nature.
The paper tackles the fragmentation of deep generative modeling frameworks by demonstrating connections between existing models and the GFlowNet framework, providing a unifying viewpoint through Markovian trajectories and a practical recipe for improvement.
There are many frameworks for deep generative modeling, each often presented with their own specific training algorithms and inference methods. Here, we demonstrate the connections between existing deep generative models and the recently introduced GFlowNet framework, a probabilistic inference machine which treats sampling as a decision-making process. This analysis sheds light on their overlapping traits and provides a unifying viewpoint through the lens of learning with Markovian trajectories. Our framework provides a means for unifying training and inference algorithms, and provides a route to shine a unifying light over many generative models. Beyond this, we provide a practical and experimentally verified recipe for improving generative modeling with insights from the GFlowNet perspective.