Analyzing Diffusion as Serial Reproduction
This work provides a novel theoretical insight into diffusion models, which are widely used in generative AI, by drawing from cognitive science, though it is incremental in nature.
The authors tackled the lack of theoretical understanding of diffusion models' properties, such as noise sensitivity and scheduling, by linking them to the cognitive science concept of serial reproduction, showing these properties emerge naturally from this correspondence.
Diffusion models are a class of generative models that learn to synthesize samples by inverting a diffusion process that gradually maps data into noise. While these models have enjoyed great success recently, a full theoretical understanding of their observed properties is still lacking, in particular, their weak sensitivity to the choice of noise family and the role of adequate scheduling of noise levels for good synthesis. By identifying a correspondence between diffusion models and a well-known paradigm in cognitive science known as serial reproduction, whereby human agents iteratively observe and reproduce stimuli from memory, we show how the aforementioned properties of diffusion models can be explained as a natural consequence of this correspondence. We then complement our theoretical analysis with simulations that exhibit these key features. Our work highlights how classic paradigms in cognitive science can shed light on state-of-the-art machine learning problems.