Neural Expectation Maximization
This addresses the challenge of identifying and manipulating conceptual entities in AI, which is crucial for tasks requiring reasoning and interaction, though it appears incremental as it builds on the Expectation Maximization framework with neural networks.
The paper tackles the problem of automated discovery of distributed symbol-like representations for tasks like reasoning and physical interaction by formalizing it as inference in a spatial mixture model with neural network components, resulting in a method that accurately recovers constituent objects in perceptual grouping tasks and learns representations useful for next-step prediction.
Many real world tasks such as reasoning and physical interaction require identification and manipulation of conceptual entities. A first step towards solving these tasks is the automated discovery of distributed symbol-like representations. In this paper, we explicitly formalize this problem as inference in a spatial mixture model where each component is parametrized by a neural network. Based on the Expectation Maximization framework we then derive a differentiable clustering method that simultaneously learns how to group and represent individual entities. We evaluate our method on the (sequential) perceptual grouping task and find that it is able to accurately recover the constituent objects. We demonstrate that the learned representations are useful for next-step prediction.