A Perspective on Objects and Systematic Generalization in Model-Based RL
This work addresses the challenge of systematic generalization in model-based RL for intelligent agents, but it is incremental as it builds on existing concepts without presenting new empirical results.
The paper argues that connectionist models do not naturally develop dynamically bound features (objects), which are crucial for modular knowledge reuse and combinatorial model construction in model-based RL, and it identifies requirements and inductive biases to address this limitation.
In order to meet the diverse challenges in solving many real-world problems, an intelligent agent has to be able to dynamically construct a model of its environment. Objects facilitate the modular reuse of prior knowledge and the combinatorial construction of such models. In this work, we argue that dynamically bound features (objects) do not simply emerge in connectionist models of the world. We identify several requirements that need to be fulfilled in overcoming this limitation and highlight corresponding inductive biases.