Active Sensing with Predictive Coding and Uncertainty Minimization
This work addresses the problem of task-independent exploration for embodied agents, offering an incremental improvement through a biologically-inspired modular approach.
The paper tackles the problem of embodied exploration by developing an end-to-end procedure inspired by predictive coding and uncertainty minimization, which discovers transition distributions and spatial features in maze navigation and builds unsupervised representations for efficient visual scene categorization. The result shows superior data efficiency and learning speed for downstream classification compared to baselines while maintaining lower parameter complexity.
We present an end-to-end procedure for embodied exploration inspired by two biological computations: predictive coding and uncertainty minimization. The procedure can be applied to exploration settings in a task-independent and intrinsically driven manner. We first demonstrate our approach in a maze navigation task and show that it can discover the underlying transition distributions and spatial features of the environment. Second, we apply our model to a more complex active vision task, where an agent actively samples its visual environment to gather information. We show that our model builds unsupervised representations through exploration that allow it to efficiently categorize visual scenes. We further show that using these representations for downstream classification leads to superior data efficiency and learning speed compared to other baselines while maintaining lower parameter complexity. Finally, the modularity of our model allows us to probe its internal mechanisms and analyze the interaction between perception and action during exploration.