Interpretable Brain-Inspired Representations Improve RL Performance on Visual Navigation Tasks
This work addresses visual navigation challenges for reinforcement learning agents, offering an interpretable method to reduce error accumulation, though it appears incremental as it builds on existing SFA techniques in a new context.
The paper tackled the problem of visual navigation by addressing the need for agents to determine their location and heading without assuming it as given, using slow feature analysis (SFA) to generate interpretable representations, resulting in improved performance where hierarchical SFA outperforms other feature extractors on navigation tasks.
Visual navigation requires a whole range of capabilities. A crucial one of these is the ability of an agent to determine its own location and heading in an environment. Prior works commonly assume this information as given, or use methods which lack a suitable inductive bias and accumulate error over time. In this work, we show how the method of slow feature analysis (SFA), inspired by neuroscience research, overcomes both limitations by generating interpretable representations of visual data that encode location and heading of an agent. We employ SFA in a modern reinforcement learning context, analyse and compare representations and illustrate where hierarchical SFA can outperform other feature extractors on navigation tasks.