Neural-Symbolic Descriptive Action Model from Images: The Search for STRIPS
This work addresses the challenge of expanding automated planning to raw image data, but it is incremental as it builds on prior neural-symbolic systems and highlights limitations rather than breakthroughs.
The paper tackled the problem of learning a descriptive PDDL action model from images for automated planning, but found that the resulting model was too large and complex for standard planners due to a translator bottleneck, achieving only competitive accuracy compared to black-box models.
Recent work on Neural-Symbolic systems that learn the discrete planning model from images has opened a promising direction for expanding the scope of Automated Planning and Scheduling to the raw, noisy data. However, previous work only partially addressed this problem, utilizing the black-box neural model as the successor generator. In this work, we propose Double-Stage Action Model Acquisition (DSAMA), a system that obtains a descriptive PDDL action model with explicit preconditions and effects over the propositional variables unsupervized-learned from images. DSAMA trains a set of Random Forest rule-based classifiers and compiles them into logical formulae in PDDL. While we obtained a competitively accurate PDDL model compared to a black-box model, we observed that the resulting PDDL is too large and complex for the state-of-the-art standard planners such as Fast Downward primarily due to the PDDL-SAS+ translator bottleneck. From this negative result, we argue that this translator bottleneck cannot be addressed just by using a different, existing rule-based learning method, and we point to the potential future directions.