Redirecting to original paper in 30 seconds...
Click below to go immediately or wait for automatic redirect
📄 Abstract
Abstract: Deep reinforcement learning (RL) agents rely on shortcut learning, preventing
them from generalizing to slightly different environments. To address this
problem, symbolic method, that use object-centric states, have been developed.
However, comparing these methods to deep agents is not fair, as these last
operate from raw pixel-based states. In this work, we instantiate the symbolic
SCoBots framework. SCoBots decompose RL tasks into intermediate, interpretable
representations, culminating in action decisions based on a comprehensible set
of object-centric relational concepts. This architecture aids in demystifying
agent decisions. By explicitly learning to extract object-centric
representations from raw states, object-centric RL, and policy distillation via
rule extraction, this work places itself within the neurosymbolic AI paradigm,
blending the strengths of neural networks with symbolic AI. We present the
first implementation of an end-to-end trained SCoBot, separately evaluate of
its components, on different Atari games. The results demonstrate the
framework's potential to create interpretable and performing RL systems, and
pave the way for future research directions in obtaining end-to-end
interpretable RL agents.
Key Contributions
This work presents the first end-to-end trained SCoBot, a neurosymbolic RL agent that decomposes tasks into interpretable, object-centric representations. This approach addresses the generalization and interpretability issues in deep RL by blending neural networks with symbolic AI, enabling demystification of agent decisions.
Business Value
Enables the development of more reliable and understandable AI agents for complex tasks, particularly in domains like robotics where safety and explainability are paramount. This can lead to faster debugging and more trustworthy AI systems.